13.11.2024. 18:15h, Faculty of Mathematics (online)
AI in Medicine: Unveiling Current Applications and Exploring New Frontiers
Dr. Ivan Lorencin
Faculty of Informatics, Juraj Dobrila University of Pula, Croatia
A meeting of the Bioinformatics seminar will be held on Wednesday, November 13th, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
LINK: https://matf.webex.com/meet/birbi
Abstract
Artificial intelligence (AI) has the potential to significantly transform medicine, enhancing diagnostic accuracy, patient care, and healthcare processes. This lecture explores the use of various AI technologies, including deep learning, large language models (LLMs), and computer vision, within medical applications. Emphasis is placed on how AI aids in the analysis of medical texts, the generation of medical reports, diagnostics from imaging, and predictive health risk assessments. By leveraging semantic segmentation, AI can precisely detect and classify tumor masses, while robotic surgery applications demonstrate improvements in precision and patient outcomes. The lecture will also address the role of AI in automating healthcare administration and accelerating drug discovery. Ethical considerations, such as transparency, data privacy, and bias prevention, are highlighted as crucial factors for trustworthy AI deployment.
Lecturer
Ivan Lorencin is an Assistant Professor at the Faculty of Informatics, Juraj Dobrila University of Pula, where he currently serves as Vice Dean for Business Cooperation and Science. He was elected to the scientific rank of Research Associate in the fields of Electrical Engineering, Mechanical Engineering, and Computer Science, and is also an editor of the journal "Mathematics." He is the author and co-author of over 100 scientific papers, more than 30 of which are published in Q1 and Q2 journals. He is actively involved in multiple AI-related projects as a project leader or team member. Besides his academic career, Ivan is also a partner and AI R&D specialist at the Croatian-Serbian startup dAIgnostics, focusing on the development of artificial intelligence solutions for medical applications, and serves as the CEO and founder of Ant Intelligence, an AI development and consulting firm.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
9.10.2024. 18:15h, Faculty of Mathematics (online)
Mapping the multiscale human
Prof. Dr. Gary Bader
The Donnelly Centre, University of Toronto, Canada
A meeting of the Bioinformatics seminar will be held on Wednesday, October 9th, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Generative models have the potential to explore how genomes encode phenotypes and biological functions. We consider the human genome as analogous to a 'natural' generative model encoding the full complexity of an individual, from cellular architecture to physiological functions. According to this perspective, generative models can provide a natural unified framework to model the human body across scales and to capture factors determining health and disease. This talk will cover examples models of parts of the human body across spatial scales and time.
Lecturer
Gary Bader is a Professor at The Donnelly Centre at the University of Toronto and an expert in Computational Biology. The Bader lab is developing computational methods and an ecosystem theory of tissue function that considers cell-cell interactions, cell growth, and cell internal mechanisms, such as pathways, reactions, and causal relationships, to help understand development, cancer and regenerative wound healing processes. See http://baderlab.org.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
15.5.2024. 18:30h, Faculty of Mathematics (online)
Machine learning in medicine: Sepsis prediction and antibiotic resistance prediction
Prof. Dr. Karsten Borgwardt
Max Planck Institute of Biochemistry, Department of Machine Learning and Systems Biology, Germany
Video: Recorded lecture (MP4, 65min, 95MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, May 15th, starting at 18:30, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Sepsis is a major cause of mortality in intensive care units around the world. If recognized early, it can often be treated successfully, but early prediction of sepsis is an extremely difficult task in clinical practice. The data wealth from intensive care units that is increasingly becoming available for research now allows to study this problem of predicting sepsis using machine learning and data mining approaches. In this talk, I will describe our efforts towards data-driven early recognition of sepsis and the related problem of antibiotic resistance prediction.
Lecturer
Karsten Borgwardt is Director of the Department of Machine Learning and Systems Biology at the Max Planck Institute of Biochemistry in Martinsried, Germany since February 2023. His work won several awards, including the 1 million Euro Krupp Award for Young Professors in 2013 and a Starting Grant 2014 from the ERC-backup scheme of the Swiss National Science Foundation. Prof. Borgwardt has been leading large national and international research consortia, including the “Personalized Swiss Sepsis Study” (2018-2023) and the subsequent National Data Stream on infection-related outcomes in Swiss ICUs (2022-2023), and two Marie Curie Innovative Training Networks on Machine Learning in Medicine (2013-2016 and 2019-2022).
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
24.4.2024. 18:15h, Faculty of Mathematics (online)
From Climate to Healthcare: Adapting Polar Diagrams for Biomedical and Machine Learning Applications
Aleksandar Anžel
Centre for Artificial Intelligence In Public Health Research, Robert Koch Institute, Berlin, Germany
Video: Recorded lecture (MP4, 52min, 62MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, April 24th, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Evaluating the performance of multiple complex models, such as those found in biology, medicine, climatology, and machine learning, using conventional approaches is often challenging when using various evaluation metrics simultaneously. The traditional approach, which relies on presenting multi-model evaluation scores in the table, presents an obstacle when determining the similarities between the models and the order of performance. By combining statistics, information theory, and data visualization, juxtaposed Taylor and Mutual Information Diagrams permit users to track and summarize the performance of one model or a collection of different models. To uncover linear and nonlinear relationships between models, users may visualize one or both charts.
This presentation will delve into the mathematical foundation of both the Taylor and Mutual Information Diagrams, highlighting their distinctions and similarities. Attendees will have the opportunity to view the resulting diagrams generated by the new library named polar-diagrams. This library offers the first publicly available implementation of the Mutual Information Diagram and the first interactive implementation of the Taylor Diagram. Furthermore, the presentation will discuss the additional features recently integrated into both diagrams, enabling the visualization of temporality or specific scalar model attributes like uncertainty. These concepts will be illustrated using sample datasets from diverse fields, including climatology, biomedicine, public health, and machine learning.
Lecturer
Dr. Aleksandar Anžel is a Postdoctoral Researcher at the Robert Koch Institute, where he actively contributes to pioneering research endeavors within the public health domain. His specialization lies in optimizing visualization techniques for AI models, developing early-detection surveillance systems, and improving existing computational workflows, among others.
Aleksandar completed his Ph.D. at Philipps-Universität Marburg, where he focused his research on improving existing and developing new bioinformatics pipelines and tools that leverage machine learning and data science methodologies. His work also revolved around developing, evaluating, and visualizing automated workflows for information storage systems utilizing molecular storage media like DNA. Moreover, he worked on improving existing and developing novel techniques for analyzing and visualizing high-dimensional multi-modal data sets, including temporal multi-omics data. During his Ph.D. Aleksandar was also employed as Technical Lead at eMedicals Healthtech GmbH, where he oversaw the development of the kidi project platform. The project, designed as Software as a Medical Device (Digital Health Application (DiGA)), was developed following rigorous industry standards to ensure patient safety and regulatory compliance.
Furthermore, Aleksandar holds a Master's degree in Mathematics with a specialization in Computer Science and Informatics from the Faculty of Mathematics at the University of Belgrade. He graduated with a Master's Thesis focusing on "Determining Protein N-glycosylation with Machine Learning Methods".
More information about Aleksandar can be found at https://aanzel.github.io/
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
10.4.2024. 18:15h, Faculty of Mathematics (online)
Using random walks to explore complex networks
Alexandre V. Morozov
Department of Physics and Astronomy and Center for Quantitative Biology, Rutgers University, USA
Video: Recorded lecture (MP4, 54min, 68MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, April 10th, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Large-scale networks represent a broad spectrum of systems in nature, science, technology, and human and animal societies. The complexity of these networks makes predictions of their properties a challenging task. I will describe a novel computational methodology, based on random walks, for the inference of both local and global properties of complex networks. I will show that our formalism yields reliable estimates of key network properties, such as its size, after only a small fraction of network nodes has been explored. Furthermore, I will introduce a novel algorithm for partitioning network nodes into non-overlapping communities - a key step in revealing network modularity and hierarchical organization. Thus, non-ergodic random walk trajectories help reveal modular organization and global structure of complex networks.
[1] Kion-Crosby, W.B. & Morozov, A.V. (2018). Rapid Bayesian inference of global network statistics using random walks. Phys. Rev. Lett. 121, 038301.
[2] Ballal, A., Kion-Crosby, W.B. & Morozov, A.V. (2022). Network community detection and clustering with random walks. Phys. Rev. Res. 4, 043117.
[3] Yu, J. & Morozov, A.V. (2024). An adaptive Bayesian approach to gradient-free global optimization. New J. Phys. 26, 023027.
Lecturer
Alexandre V. Morozov received his Ph.D. in Physics from the University of Washington, Seattle in 2003. From 2003 to 2007 he was a post-doctoral fellow at the Center for Studies in Physics and Biology, Rockefeller University, New York. In 2007 he joined the Department of Physics and Astronomy at Rutgers University, where he is now Professor and Director of the Center for Quantitative Biology. In 2009, he was a recipient of an Alfred P. Sloan Research Fellowship. His current research interests include non-equilibrium statistical mechanics, machine learning, biological physics, and evolutionary modeling.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
27.3.2024. 18:15h, Faculty of Mathematics (online)
Computational image analysis of cell nuclei textures in pediatric patients suffering from inflammatory bowel disease
Vedrana Makević, Dr. Anđelija Ilić
Faculty of Medicine, University of Belgrade, Serbia
Institute of Physics, University of Belgrade, Serbia
Video: Recorded lecture (MP4, 60min, 101MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, March 27th, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
We will present our recent research results obtained in collaboration of Faculty of Medicine, University of Belgrade (UB), and Institute of Physics Belgrade, UB, with the Department of Gastroenterology, Hepatology, and Endoscopy, University Children’s Hospital, UB, namely Dr. Ivan Milovanovich, MD, PhD, and N. Popovac, MD. We focused on the inflammatory bowel disease (IBD) in pediatric population. The incidence of IBD has been in increase in recent years, with 20–30% of cases beginning in pediatric age. While an early diagnosis is essential for obtaining timely treatment and preventing more permanent consequences, it can be very challenging. In particular, discerning between the two different disease phenotypes, Crohn’s disease (CD) and ulcerative colitis (UC), often presents a problem even to the experienced pathologists. We complemented standard biochemical and immunohistochemical analyses with the extraction of morphological and textural features from the high-resolution digitalized micrographs of intestinal cell nuclei. The aim was to check out the possibility to discern the IBD patients from the healthy controls based on some of the studied parameters, as well as to hopefully find the combinations of parameters which could help differentiate between the Crohn’s disease and ulcerative colitis. Our motivation to use the intestinal cell nuclei extracted from the crypt glandular epithelium arose from the prior investigations of our own as well as of other groups, showing significantly altered textural features of cell nuclei chromatin in cancers. Given the role of genetic, immune, and environmental factors in IBD, we presumed that the analysis of intestinal cell nuclei might show the existence of modified cell nuclei textures in IBD. Each of the seven distinct intestinal tissues was considered separately and in addition to the intergroup comparisons, we also showed intersegmental comparisons of the selected cell nuclei features. Very promising results, in agreement and additionally supporting other analyses were obtained. Some of the observed correlations with other studied manifestations of IBD will be discussed. The conducted comprehensive investigation showed some promise of the use of the proposed methodology as one of the constituent parts in the development of computer assisted IBD diagnosis tools.
Lecturers
Vedrana Makević was born in Belgrade, Serbia, on July 7th 1988. She received the MD degree in 2013 from the Faculty of Medicine, University of Belgrade (MFUB). Since 2014, she is enrolled into the MFUB graduate school program on Physiological sciences, towards obtaining her PhD degree. During the same period, she was employed at the Institute of Pathological Physiology at MFUB, where she currently works. Since 2022, she took part in the currently ongoing project of the Science Fund of the Republic of Serbia, program IDEAS, grant No. 7673781, “Polyphenols as potential targeted treatments in D. melanogaster model of fragile X syndrome”, as well as (since 2023.) in the ERASMUS+ project “Development and implementation of metacognitive problem-based modules in blended learning courses in medical sciences, ProBLeMS”. Within her PhD research, under the guidance of Prof. Dr. Silvio de Luka (MFUB) and Dr. Andjelija Ilić (Inst. of Physics Belgrade, UB), she studied pediatric IBD cases, combining standard methodology with innovative methods such as determining metal and trace element contents in intestinal tissues and conducting digital image analyses of intestinal cell nuclei morphology and texture.
Dr. Andjelija Ilić was born in Belgrade, Serbia, in 1973. She received the Dipl. Eng., M.Sc., and Ph.D. degrees in electrical engineering in 1998, 2004, and 2010, from the University of Belgrade (Belgrade, Serbia), the University of Massachusetts Dartmouth (MA, USA), and the University of Belgrade (Belgrade, Serbia), respectively. She is currently an Associate Research Professor with the Institute of Physics, University of Belgrade. In 2013 and 2014, she was a Postdoctoral Research Associate with the School of Science and Technology, University of Westminster, London, UK. Her research interests are in the fields of applied physics and electromagnetics, including numerical methods, microwave circuits, accelerator technology, application of novel materials in electrical engineering, and electromagnetic field interaction with biological systems. Since 2012, she collaborates with MFUB, in diverse studies including the biomedical image analysis focused onto specific problems in medical sciences, which can be challenging in practice, and could therefore benefit mostly from the accurate and quantitative analysis.
She holds two patents in biomedical engineering. She was the principal investigator of the Innovation project for 2018–2019, No. 391-00-16/2017-16/27, “Development of a new type of device for electroporation of cells and tissues using ultra short electric pulses”, financed by the Ministry of Education, Science, and Technological Development of the Republic of Serbia. Currently, she participates in two projects funded by the Science Fund of the Republic of Serbia, Green project No. 5661, IonCleanTech, “Elimination of respirable airborne particles, microplastics, microorganisms, and VOCs by ionization of indoor air and filtration systems: comprehensive investigation for reliable technological answers”, and PRISMA project No. 7328, ToxoReTREAT, “Reinvention of the diagnostic algorithm and treatment options for reactivated toxoplasmosis”.
Dr. Ilić is the Senior Member of IEEE. She was the recipient of the 2006 Young Scientist and the 2014 Best Paper ETRAN Awards, as well as the “Prof. Aleksandar Marinčić” Award given annually by the IEEE MTTS Serbia chapter, for the best journal paper in 2016.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
21.2.2024. 18:15h, Faculty of Mathematics (online)
Decision Support Systems for Early Detection of Alzheimer's Disease
Teodora Srećković, Milica Vukašinović
School of Electrical Engineering, University of Belgrade
A meeting of the Bioinformatics seminar will be held on Wednesday, February 21st, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
According to the latest report from Alzheimer's Disease International, nearly 55 million people worldwide are living with dementia. The World Health Organization estimates the number will triple by 2050, along with encouragement that timely diagnosis, early therapy treatments, and lifestyle changes can significantly postpone the further progression of the disease. In this talk will be presented research on decision support systems in healthcare that can aid early detection of Alzheimer’s disease, the most common form of dementia. The research is based on complementary techniques of Natural Language Processing and Computer Vision and conducted on publicly available datasets.
Lecturers
Teodora Srećković and Milica Vukašinović are both fourth-year students at the School of Electrical Engineering, University of Belgrade. The research on decision support systems in healthcare was conducted as part of the Summer Internship Program in 2023 at the Mathematical Institute of the Serbian Academy of Sciences and Arts, which is well-known for its academic excellence. They explored the capacity of the latest techniques of Natural Language Processing and Computer Vision in the field of dementia under the supervision of Anđelka Zečević. The results were also presented in the seminar Decision Making - covering the theory, technology, and practical application at the Mathematical Institute.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
13.12.2023. 18:15h, Faculty of Mathematics (online)
Biomedicine in the Age of Generative AI
James Zou
Stanford University
Video: Recorded lecture (MP4, 63min, 77MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, December 13th, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
There have been tremendous advances in generative AI such as ChatGPT and DALLE. This talk will explore how generative models can power biomedical discoveries by expanding the design space of medicine while balancing complex tradeoffs. I will illustrate this through three examples. We will first discuss how to use generative AI to design and experimentally validate novel drugs. Then we will discuss how to use large language models to help us build foundation models for cell biology. Finally, we will demonstrate how to build (using Twitter!) visual-language models to index complex biomedical data.
Lecturer
James Zou is an assistant professor of Biomedical Data Science, CS and EE at Stanford University. He is also the faculty director of Stanford AI4Health. He works on both improving the foundations of ML–-by making models more trustworthy and reliable–-as well as in-depth scientific and clinical applications. Many of his innovations are widely used in tech and biotech industries. He has received a Sloan Fellowship, an NSF CAREER Award, two Chan-Zuckerberg Investigator Awards, a Top Ten Clinical Achievement Award, several best paper awards, and faculty awards from Google, Amazon, Tencent and Adobe. His research has also been profiled in popular press including the NY Times, WSJ, and WIRED.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
29.11.2023. 18:15h, Faculty of Mathematics (online)
Introduction to Computational Neuroscience: The Hodgkin-Huxley model
Lea Kojičić
University of Belgrade, Faculty of Mathematics
Video: Recorded lecture (MP4, 61min, 75MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, November 29th, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Computational neuroscience (CNS) is a multidisciplinary field, with the aim to explain the principles that govern neural systems using quantitative models. These models are used to generate hypotheses about the mechanisms underlying observed data, which lead to new discoveries about the phenomenon of interest. This presentation introduces CNS through its pioneering work - the Hodgkin Huxley (HH) model.
The HH model explains initiation and propagation of action potential in the axon of a giant squid. It introduced the idea that the permeability of the membrane to different ions changes dynamically during an action potential, leading to the novel concept of ion channels. We first reflect on the process and importance of cell excitation, and move forward to the centerpiece model. After a brief history of the research, we will derive all the equations the model consists of, and discuss their importance in modern neuroscience.
Lecturer
Lea Kojičić is a third year student at the Faculty of Mathematics, University of Belgrade. She wishes to couple her passion for mathematics and a long lasting fascination with the brain, by pursuing a MSc degree in computational neuroscience. In her initiation into the field, she authored a high school final year seminar paper on the Human Hypothalamus, earning honorable mentions from Mathematical Grammar School. As a sophomore, Lea got familiar with processing and understanding EEG signals at the Institute for Medical Research in Belgrade, where she did a research internship. Primarily self-taught, she actively engaged within the field by undertaking advanced mathematics courses at her university over the past two years, which helped her understand various research papers and Computational Neuroscience books. As of this semester, she is enrolled in the Neurobiology course at the Faculty of Biology in Belgrade. Her current interests include Affective Computational Neuroscience and Brain to Brain Interfaces.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
15.11.2023. 18:15h, Faculty of Mathematics (online)
Challenges in variant interpretation for rare disease diagnosis
Anne O'Donnell Luria, M.D., Ph.D.
Broad Institute of MIT and Harvard
Video: Recorded lecture (MP4, 68min, 85MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, November 15th, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Rare genetic disorders, while individually rare, are collectively common. Over the last decade, advances in genomic methods have significantly uplifted diagnostic rates for patients and facilitated novel and targeted therapies. However, >60% of patients with rare genetic disorders remain undiagnosed. One approach to improving rare disease diagnosis is through the use of human population data, demonstrating what variation is seen in the general population and also what expected variation is missing. The recent release of the Genome Aggregation Database (gnomAD) v4 with exome or genome data from >800,000 individuals including >150,000 from non-European ancestries provides a large catalog of human genetic variation. I will discuss how we use this dataset in rare disease diagnosis and areas that remain challenging in identifying functional variation.
Lecturer
Anne O'Donnell-Luria is co-director of the Center for Mendelian Genomics at the Broad Institute of MIT and Harvard and an Assistant Professor in Pediatrics at Boston Children’s Hospital, Harvard Medical School. Her research focuses on using large-scale genomic and transcriptomic approaches to increasing the rate of rare disease diagnosis through improving rare variant interpretation and empowering the discovery of novel disease genes. Through the Broad Institute Center for Mendelian Genomics and GREGoR consortium, she has contributed to diagnosing several thousand patients and several hundred gene-disease relationship discoveries. She is particularly interested in how we can leverage large reference population databases such as gnomAD to improve diagnosis and estimating rare disease prevalence. She is committed to improving standards for variant and gene classification through her roles in ClinGen including co-chairing the Syndromic Disorders Gene Curation Expert Panel. Her clinical genetic focus is in caring for children with disorders involving the chromatin machinery, including Kleefstra syndrome and KMT2E-neurodevelopmental condition.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
18.10.2023. 18:15h, Matematički fakultet (on ground and online)
Uncertainty quantification and neural network interpretation for studying CRISPR mechanics
Bogdan Kirillov, Ph.D.
Skolkovo Institute of Science and Technology, Moscow
Institute of Gene Biology, Russian Academy of Sciences
A meeting of the Bioinformatics seminar will be held on Wednesday, October 18th, starting at 18:15, in classrom 718 and in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
CRISPR-Cas genome editing is a powerful tool used for both basic biological research and industrial biotechnology applications. Although the process of using CRISPR-Cas proteins to accurately cleave the target DNA might seem straightforward, the experimental planning actually involves numerous steps, each adding noise to the data, requiring advanced mathematical techniques for analysis. This talk focuses on software tools for biotechnological applications of CRISPR-Cas that use Deep Neural Networks, Explainable Machine Learning, and Uncertainty Quantification. In this presentation, two projects that investigated potential off-target event detection and cleavage efficiency estimation are discussed. Due to their superior accuracy and efficiency, these methods are a valuable addition to the toolkit of experimental biologists. In particular, one project introduces a general method for anomaly detection that passed a careful comparison study, and another one introduces a measure for prediction variance that allows for studying previously unknown dimension of potential off-target events in Cas9/Cas12a-mediated gene editing. Additionally, the performance of cleavage efficiency estimation received an independent biological validation through rediscovery of known Cas-protein behavior without any kind of supervision. The presentation also discusses a case study for gene editing experiment design using the aforementioned tools.
Lecturer
Dr. Bogdan Kirillov is a bioinformatician with both an academic background and practical industry experience. He earned his B.Sc. in Applied Physics from ITMO University and went on to get his M.Sc. and PhD in Life Sciences from the Skolkovo Institute of Science and Technology. Currently, Dr. Kirillov is part of the Center for Precision Genome Editing and Genetic Technologies for Biomedicine at the Institute of Gene Biology, Russian Academy of Sciences. In the industry, Dr. Kirillov has worked as a Machine Learning Engineer for several Russian IT companies, including SberMedAI, a medical AI subsidiary of Sber (the largest Russian bank). Dr. Kirillov's research interests span a broad spectrum within the field of Computational and Synthetic Biology. He is particularly intrigued by the application of machine learning in bioinformatics, the exploration of CRISPR-Cas systems, and the innovative field of 3D bioprinting.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Nataša Pržulj and dr Jovana Kovačević.
10.5.2023. 18:15h, Matematički fakultet (online)
Semantic Unification and Search of Bioinformatics Databases
Aleksandar Veljković
Video: Recorded lecture (MP4, 49min, 54MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, May 10th, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Associating biological data from different sources provides a holistic view of a domain and enables finding patterns in data that are otherwise difficult or impossible to observe by only analyzing isolated biological entities. The key issues for creating connections between data objects are the variety of biological data formats, data organization schemas, and data access methods. Connecting data from different databases can be challenging, as an entity from one database may not have the same properties or identifiers as the same entity described in another database. While some databases contain a variety of entity identifiers from various databases, the search is limited to exact property matching, and complex queries using multiple metadata attributes are not possible. To overcome these issues, a novel data framework BioGraph enables linking and retrieving information from heterogeneous interconnected biological data. The model was tested and generated a knowledge graph using metadata from five distinct public datasets, DisProt, HGNC, Tantigen 2.0, IEDB, and DisGeNET. The resulting graph interconnects more than 17 million nodes, of which 2.5 million individual biological entity objects with over 4 million relationships. The software system allows searching and retrieving patterns and retrieving matching results from the knowledge graph using a user-friendly interface. To complement the model, a tool and a web interface were developed. The tool and corresponding packages can be deployed locally as a standalone system, enabling offline execution of queries.
Lecturer
Aleksandar Veljković is a Ph.D. student and teaching assistant at the Faculty of Mathematics, University of Belgrade. He is a member of the bioinformatics research group at the Faculty of Mathematics and treasurer of the BIRBI organization. His domains of research include data science, bioinformatics, and cryptography.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
19.4.2023. 18:15h, Matematički fakultet (online),
Multi-view boosting with adversarial multi-arm bandits on incomplete microbiome views
Andrea Simeon (born Mihajlović)
Video: Recorded lecture (MP4, 39min, 43MB)
Presentation: Download (PDF, 2.4MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, April 19th, starting at 18:15, in classrom 718 and in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Microbiome has been massively associated with different diseases and disorders. To identify individual microorganisms and their abundances across samples, different sampling, sequencing and preprocessing techniques could be considered. This leads to different input feature sets (views) to learn predictive models through machine learning (ML) approaches. ML models aid in finding the associations between microbiome and disease. Standard (single view) ML models are not capable of dealing with multiple views at once, and thus they were upgraded to fit multi-view datasets (e.g. Adaboost and Multi-view Adaboost). Moreover, microbiome data comes from various sources and often view incompleteness is inevitable. Existing classifiers, even multi-view, cannot be directly used because they cannot work with incomplete views and in multi-class settings. To the best of our knowledge, there is no multi-view boosting algorithm for multi-class classification with incomplete views.
The proposed algorithm is the extension of an existing multi-view boosting algorithm based on multi-arm bandits, now able to work in multi-class setting and with incomplete views (views with missing sample representation). At each iteration, it proclaims one view as the winning using adversarial multi-arm bandits and uses its predictive information to update the final model weights and prediction in a boosting process. Three data sets were created from several microbiome studies and used to examine the performance of the proposed algorithm. One of the experiments showed a 7% increase in F1 score compared to a single view classifier, while the other one showed 54%. The application domain is not restricted to microbiome data. Further work will involve examinations in other domains.
Lecturer
Andrea Mihajlovic is a Junior Research Assistant at BioSense Institute, and PhD student in Computer Science, Faculty of Sciences, both at University of Novi Sad, Serbia, with a background in applied mathematics (BSc) and data science (MSc). She is interested in digitalization and Artificial Intelligence (AI) for improved health and disease status assessment based on diverse omics data. Currently, she is focused on applying AI techniques in microbiome studies and exploring different preprocessing pipelines for analyzing amplicon and shotgun sequence data.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
29.3.2023. 18:15h, Matematički fakultet (online),
Unlocking fertility in women via human omics data
Dr Staša Stanković
Video: Recorded lecture (MP4, 62min, 73MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, March 29, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Menopause timing is highly variable, having a direct effect on reproductive lifespan, fertility and health outcomes in later life. Endocrine and imaging tests only record changes in ovarian function that have already taken place, thus disabling early prediction and identification of women with reduced reproductive lifespan, such as primary ovarian insufficiency (menopause below 40). Human genetic studies have attempted to overcome this problem by identifying genetic markers associated with menopause timing and fertility. Using data from large scale population studies, including UK Biobank, 23&Me and deCODE Genetics, we assessed both common and rare genetic variation that influence menopause timing in women. Our work on common genetic variants led to the discovery of over 300 genetic signals (Nature, 2021) that influence the age women begin menopause and the first evidence of our ability to, through gene manipulation in a mouse model, extend reproductive lifespan by 25% and improve fertility. The power of this information is that in the future we may be able to build the first genetic prediction test that will inform every woman about the timing of her menopause. In addition, these findings are critical as improved knowledge of the underlying mechanisms may also allow their manipulation, more specifically halting or temporizing the process of the loss of ovarian follicles and provide a new direction for therapeutic approaches that might seek to treat infertility. To assess the impact of rare damaging variants on age at natural menopause, we queried whole-exome sequencing data for 106,973 post-menopausal women in UK Biobank and implicated novel genes with effect sizes up to 6 times larger than previously discovered (under review, Nature). Finally, we found that genetic susceptibility to earlier ovarian ageing in women increases de novo mutation rate in their offspring. This provides direct evidence that female mutation rate is heritable and highlights a mechanism of the maternal genome influencing child health, which could have direct implications for the health of future generations given the link between de novo mutations and disease risk. Our study provides biological insights into reproductive ageing by increasing the number of implicated genes with a potential to inform experimental studies seeking to identify new therapies that enhance reproductive function and preserve fertility.
Lecturer
Dr Staša Stanković is a geneticist and bioinformatician with a PhD in Reproductive Genomics from the University of Cambridge. Her work is focused on deciphering the genetic architecture of reproductive ageing and fertility using large-scale population omics data, and their link to later life health outcomes in women. The work by Staša and her collaborators led to the discovery of genetic signals that influence the age women begin menopause and the first evidence of the ability to, through gene manipulation in mouse model, extend reproductive lifespan and improve fertility. Staša and her collaborators are using these ground-breaking findings to embark on the commercialisation journey towards the development of prediction tests and next generation therapeutics for reproductive disorders.
Prior to her PhD, Staša was awarded an MPhil in Medical Science from the University of Cambridge, Wellcome Trust Sanger Institute and dual BSc in Biomedical Sciences from the University of Oxford and Oxford Brookes University. Staša was also part of Congenica, genomic data analysis company, where she contributed to the development of “Congenica Neuro”, new generation product for the tailored genomic analysis of individuals with neurodevelopmental diseases, enabling clinicians to provide >20X faster diagnosis.
She was a founder of Innovation Forum Serbia with an aim to build a mechanism that will support the generation and development of Serbian healthcare startup ecosystem, and bridge the gap between academia, industry, policy makers and investors. In the ecosystem where this concept was unexplored at the time, she managed to attract strategic and financial support by leading governmental (UK and Serbian Government, NHS) and industrial (Roche, AstraZeneca, Microsoft etc.) institutions to pioneer crucial systematic changes. This enabled her to establish the first ever healthcare accelerator programme in Serbia, ‘IMAGINE IF!’ in collaboration with Science Technology Park Belgrade.
She is also a business consultant, where she built numerous digitalisation and innovation strategies for major pharmaceutical and governmental institutions. Based on her impact, she was awarded ‘McKinsey&Company Next Generation Women Leader’ for 2020, the best young investigator under the age of 40 by the International Menopause Society ('Robert Greenblatt Award'), one of ten best UK young scientists in the biomedical sector by the UK Government ‘STEM for Britain’, and she is considered as one of 11 scientists with the highest impact on innovations in menopause by Forbes.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
15.3.2023. 18:15h, Matematički fakultet (on ground and online, in Serbian)
C4IR in Serbia: Storage and analysis of genomic data
Dr Branislava Gemović
Presentation: Download (PPTX, 4.5MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, March 15, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
This lecture will be in Serbian, on ground and online.
Abstract
The Centre for the Fourth Industrial Revolution (C4IR) in Serbia was established in February 2022, by the Government of the Republic of Serbia and the World Economic Forum, with two major fields of activities – biotechnology and the application of artificial intelligence in healthcare. The Centre is a part of the Office for IT and eGovernment and, as such, is involved in the building and strengthening national infrastructure resources for storage and analysis of the data such as genetic and biomedical data.
C4IR is coordinating two projects focused on genomic data and this talk will cover some of the activities. Current stage of development of the Platform for storage and sharing of genetic and biomedical data in Serbia will be presented, with a brief overview of the functional specification of the informational system. Secondly, a pilot project – DNA screening: sequencing for the prevention of cardiovascular and oncological diseases will be described and discussed. Both projects and other activities of the Centre greatly involve Serbian bioinformatics community and strengthening this community is one of the important roles of the Centre, therefore the discussion about the strategy for this is one of the aims of this lecture.
This lecture will be in Serbian, on ground and online.
Lecturer
Branislava Gemović is a project manager at the Centre for the Fourth Industrial Revolution, in charge of preparation and coordination of projects in the field of artificial intelligence and biotechnology. She is also employed at the Vinča Institute of Nuclear Sciences, University of Belgrade, and as a lecturer at the Faculty of Stomatology Pančevo.
Branislava has a Ph.D. in biology and more than ten years of experience in scientific research in the field of bioinformatics and computational biology. She has participated in a number of research efforts related to the use of machine learning in biomedicine, resulting in bioinformatics software tools. At the moment, her major focus is on the development of the Platform for storage and sharing of genetic and biomedical data in Serbia.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
22.2.2023. 18:15h, Matematički fakultet
SemiBin - Using self-supervised deep learning for better metagenomics binning
Dr Luis Pedro Coelho
A meeting of the Bioinformatics seminar will be held on Wednesday, February 22, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Microorganisms live in all environments on the globe and influence phenomena as varied as human health, soil (including soil used for food production), and climate change. Metagenomics is the technique whereby DNA sequencing is simultaneously performed on the multiple organisms present in a particular sample. Due to technological limitations, metagenomics returns a myriad of fragments, often named "contigs", rather than complete genomes. Attempting to infer which contigs form genomes is called binning.
The two most popular approaches, reference-based and reference-independent binning map to the machine learning concepts of supervised and unsupervised learning. Reference-based binning can work well to discover variations of already-known species, but only reference-free methods can find novel ones and a large fraction of contigs cannot be classified into known species (in many cases, these are the vast majority of contigs in a sample). SemiBin originated as an attempt to bridge the gap between these two approaches and develop a semi-unsupervised model in which supervision was used to inform the process, but it was still possible to infer novel groups. While successful at recovering more genomes than the previous state-of-the-art, this approach was computationally intensive. Recently, we have overcome this by developing a completely self-supervised approach, which achieves the high quality results of the semi-unsupervised model (even surpassing it) at a fraction of the computational costs.
Additionally, we have made the pipeline work with either the short-read datasets that still comprise the majority of metagenomics as well as the newer long-read methods that are increasingly used. Although the problem of clustering short or long-read data are conceptually identical, their properties are sufficiently distinct that we found that using different approaches for each achieves the best results.
SemiBin is available at https://github.com/BigDataBiology/SemiBin and the manuscript describing the semi-supervised version is available at https://doi.org/10.1038/s41467-022-29843-y A preprint describing the recent updates is available at https://www.biorxiv.org/content/10.1101/2023.01.09.523201v1
Lecturers
Luis Pedro Coelho is the principal investigator (PI) of the Big Data Biology Lab at Fudan University. Previously, he worked as a postdoctoral researcher in Peer Bork's group at the European Molecular Biology Laboratory (EMBL). He has a PhD from Carnegie Mellon University and an MSc from Instituto Superior Técnico in Lisbon. He currently works on the analysis of microbial communities in different environments, such as the marine environment or the human gut using computational methods.
More information about can be found at https://luispedro.org.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
8.2.2023. 18:15h, Matematički fakultet
Machine intelligence and network science for complex systems big data analysis
Prof. Dr Carlo Vittorio Cannistraci
Video: Recorded lecture (MP4, 83min, 123MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, February 8, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
I will present our research at the Center for Complex Network Intelligence (CCNI) that I recently established in the Tsinghua Laboratory of Brain and Intelligence at the Tsinghua University in Beijing. We adopt a transdisciplinary approach integrating information theory, machine learning and network science to investigate the physics of adaptive complex networked systems at different scales, from molecules to ecological and social systems, with a particular attention to biology and medicine, and a new emerging interest for the analysis of complex big data in social and economic science. Our theoretical effort is to translate advanced mathematical paradigms typically adopted in theoretical physics (such as topology, network and manifold theory) to characterize many-body interactions in complex systems. We apply the theoretical frameworks we invent in the mission to develop computational tools for machine intelligent systems and network analysis. In particular, we deal with: prediction of wiring in networks, sparse deep learning, network geometry and multiscale-combinatorial marker design for quantification of topological modifications in complex networks. This talk will focus on two main theoretical innovation. Firstly, the development of machine learning for topological estimation of nonlinear relations in high-dimensional data (or in complex networks) and its relevance for applications in big data, with a particular emphasis on brain connectome analysis. Secondly, we will discuss the Local Community Paradigm (LCP) and its recent extension to the Cannistraci-Hebb network automata, which are brain-inspired theories proposed to model local-topology-dependent link-growth in complex networks and therefore are useful to devise topological methods for link prediction in sparse deep learning, or monopartite and bipartite networks, such as molecular drug-target interactions and product-consumer networks.
Lecturers
Carlo Vittorio Cannistraci is a theoretical engineer, Zhou Yahui Chair Professor, Chief Scientist at the Tsinghua Laboratory of Brain and Intelligence (THBI), Director of the Center for Complex Network Intelligence (CCNI) at THBI and member of the Department of Computer Science and the Department of Biomedical Engineering at the Tsinghua University, Beijing, China. Carlo’s area of research embraces information theory, machine learning and physics of complex systems and networks including also applications in systems biomedicine and neuroscience. Nature Biotechnology selected Carlo’s article (Cell 2010) on machine learning in developmental biology to be nominated in the list of 2010 notable breakthroughs in computational biology. Circulation Research featured Carlo’s work (Circulation Research 2012) on leveraging a cardiovascular systems biology strategy to predict future outcomes in heart attacks, commenting: “a space-aged evaluation using computational biology”. The Technical University Dresden honoured Carlo of the Young Investigator Award 2016 in Physics for his work on the local-community-paradigm theory and link prediction in monopartite and bipartite networks. In 2017, Springer-Nature scientific blog highlighted with an interview to Carlo his study on “How the brain handles pain through the lens of network science”. The American Heart Association covered this year on its website the recent chronobiology discovery of Carlo on how the sunshine affects the risk and time onset of heart attack. In 2018, Nature Communications featured Carlo’s article entitled “Machine learning meets complex networks via coalescent embedding in the hyperbolic space” in the selected interdisciplinary collection of recent research on complex systems. In 2019, Scientific Reports selected Carlo’s interview between all their Editors to represent the journal in the social media. In 2019, Carlo won the Shanghai 1000 talents plan award, sponsored by CAS-MPG Partner Institute for Computational Biology. In 2020, Carlo was awarded of the Zhou Yahui Chair Professorship of Tsinghua University. In 2021, Carlo’s won the National high-level talent program award from the Minister of Science of China.
References: Google Scholar
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
14.12.2022. 18:15h, Matematički fakultet (on ground and online)
1. Center for Genomic Sequencing and Bioinformatics: current and future perspectives
2. Genome wide association study of COVID-19 clinical outcomes in population of Serbia
dr Mirjana Novković, Marko Zečević
Video: Recorded lecture (MP4, 49min, 54MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, December 14, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
1. Center for Genomic Sequencing and Bioinformatics: current and future perspectives
High throughput technologies become essential tools in molecular biology allowing the fast analysis of large biological data in just a few days. IMGGE recently established the Centre for genome sequencing and bioinformatics which operates state of the art resources for next generation sequencing (NGS) including MGI, Illumina and Oxford Nanopore platforms, as well as mass spectrometry instrument for proteomic analyses. The Center also includes space and infrastructure for sequencing data analysis, with local storage servers and additional access to external servers such as server at Faculty of Physics and Nvidia as a part of National Data Center Office. Resources of the Center are currently used for Sars-Cov-2 genome sequencing, prenatal genome testing and whole genome and whole exome sequencing. One of the main goals of the newly established Center also includes developing and automatizing NGS data analysis pipelines that will be made available to the scientific community which will help researchers less experienced in bioinformatics to analyze their NGS data more efficiently.
2. Genome wide association study of COVID-19 clinical outcomes in population of Serbia
Host genetics, an important contributor to the COVID-19 clinical susceptibility and severity, currently is the focus of multiple genome-wide association studies (GWAS) in populations affected by the pandemic. This is the first study from Serbia that performed a GWAS of COVID-19 outcomes to identify genetic risk markers of disease severity. A group of 128 hospitalized COVID-19 patients from the Serbian population was enrolled in the study. A GWAS was conducted comparing (1) patients with pneumonia (n = 80) against patients without pneumonia (n = 48), and (2) severe (n = 34) against mild disease (n = 48) patients, using a genotyping array followed by imputation of missing genotypes. The previously reported COVID-19 risk locus at 3p21.31 was replicated - suggestively associated in our data with both pneumonia and severe COVID-19. A significant signal associated with COVID-19-related pneumonia was detected at locus 13q21.33 and several suggestive associations have also been observed at chromosomes 5p15.33, 5q11.2, and 9p23. The genes located in or near the risk loci are expressed in neural or lung tissues, and have been previously associated with respiratory diseases such as asthma and COVID-19 or reported as differentially expressed in COVID-19 gene expression profiling studies. Our results revealed novel risk loci for pneumonia and severe COVID-19 disease which could contribute to a better understanding of the COVID-19 host genetics in different populations.
Lecturers
Mirjana Novković is assistant research professor at Institute of Molecular Genetics and Genetic Engineering (IMGGE) University of Belgrade, Serbia. For the past few years she attended different bioinformatics courses including program "Bioinformatics for Biologists" at School of Computing, University Union in Belgrade. She is mostly dedicated to whole genome and whole exome sequencing data analyses and interested in transcriptomic data analysis. Currently she is team leader for Sars-Cov-2 genome sequencing at the Center for Genome sequencing and Bioinformatics at IMGGE.
For the past seven years, Marko Zečević has been working as a bioinformatics analyst at Seven Bridges, the industry-leading bioinformatics ecosystem provider. He is currently pursuing a Ph.D. in genomics, investigating genetic risk factors for severe COVID-19 in the Laboratory for Molecular Biomedicine at the Institute of Molecular Genetics and Genetic Engineering (IMGGE) in Belgrade, Serbia.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
23.11.2022. 18:15h, Matematički fakultet (on ground and online)
Hurst Space Analysis – data discrimination tool based on cyclic trends in records
dr Suzana Blesić
A meeting of the Bioinformatics seminar will be held on Wednesday, November 23, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
It was shown for variables across different complex systems that their fluctuation functions calculated with detrending methods of scaling analysis frequently exhibit existence of transient crossovers in behavior, signs of trends that arise as effects of periodic or aperiodic cycles. We recently developed a technique to cluster or differentiate records from an arbitrary complex system dataset based on the presence and influence of these cycles in data, which we dubbed the Hurst Space Analysis (HSA). We defined a space of -vectors that represent records in the dataset, which we called the Hurts space. Vectors are populated by scaling exponents calculated on subsets of time scale windows of time series that bound cyclic peaks in their global wavelet power spectra (WTSs), by way of use of time dependent detrended moving average analysis (tdDMA). The length depends on the number of WTS peaks in that complex system. This number is, as was shown across complex systems, universal. To be able to quantify any time series with a single number, we projected their relative Hurst space unit vectors (with ) onto a unit vector of an assigned preferred direction in the Hurst space. The definition of the ’preferred’ direction depends on the characteristic behavior one wants to investigate with HSA - projection of unit vectors of any record with a ’preferred’ behavior onto the unit vector will then always be positive.
The HSA procedure can serve to examine and differentiate records within datasets of randomly selected time series of any complex variable. We used HSA to differentiate complex time series of stock market data, based on the preferred characteristic of marked development, and to cluster datasets of observed temperature records from land stations from different climatically and topologically homogeneous regions, based on the ‘belonging to a continent’ preference.
Lecturer
Suzana Blesić is an assistant research professor at the Institute for Medical Research in Belgrade. She holds a PhD in theoretical statistical physics, a post-doc at CNRS Marseilles, has worked in laboratories in Sweden and Japan, and has led a Marie Curie research project at the Ca’Foscari University of Venice in Italy. She has explored disparate research fields - first neuroscience, then finance and now climate change - with two purposes: analyzing and understanding complex systems and developing methodological frameworks for (their) data analysis. Currently she leads a Group for Biomechanics, Biomedical Engineering and Physics of Complex Systems at her home institute. She is also a co-PI and a WP leader of the Horizon Europe project that will investigate relationships between vector borne diseases and climate and environmental change and develop preparedness tools for adaptation to that change.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
02.11.2022. 18:15h, Matematički fakultet
Challenges in Metagenomic Annotation of Antibiotic Resistome
dr Svetlana Ugarčina Perović
Video: Recorded lecture (MP4, 37min, 46MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, November 2, starting at 18:15, in online classroom. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
The antibiotic resistance genes (ARGs) in both host-associated and environmental microbiomes – antibiotic resistome – play an important role in the spread of antibiotic resistance. Metagenomics enables high-throughput exploration of microbiomes. One important step in these analyses is ARGs annotation within the metagenomes, for which several tools are in use. However, most of these tools are developed for genomics studies and their databases may pose certain biases. We aimed to compare outputs from different ARGs annotation tools for metagenomes and detect their potential challenges in microbiome studies. We ran >13 000 high-quality metagenomes from 14 habitats (Coelho et al., 2022; https://gmgc.embl.de/) through three ARGs annotation pipelines (with default settings) using DeepARG (CARD, ARDB, UNIPROT; https://bench.cs.vt.edu/deeparg), RGI (CARD; https://card.mcmaster.ca/) and ABRicate (CARD, ResFinder, ResFinderFG, NCBI, MEGARes, ARG-ANNOT; https://github.com/tseemann/abricate), and compared their outputs ( https://github.com/pha4ge/hAMRonization ). To facilitate comparison of outputs with different gene_names, we performed ARO normalization (https://github.com/AdeBC/quick_amr_db_harmonisation, based on work by Finlay Maguire) i.e. mapping NCBI, ResFinder and ResFinderFG to CARD's ARO ontology. DeepARG and RGI provide higher coverage (with potential novel ARGs and/or false positives) while ABRicate with different databases has lower coverage but more well validated ARGs. The annotations did not differ only in number of hits but also in the information provided: e.g.using ABRicate, more sulfonamide resistance genes were identified using ResFinderFG version 2.0 as the database (ARGs obtained by functional metagenomics by Gschwind et al.), while using ResFinder resulted in more macrolide-resistance genes. Thus, choice of annotation tool and database should be driven by research questions and ARGs targets.
Lecturer
Dr Svetlana Ugarčina Perović is a postdoctoral researcher in the Big Data Biology Lab at Fudan University (Shanghai, China) https://www.big-data-biology.org/. Svetlana holds a B.S. and Ph.D. in Environmental Sciences from the University of Novi Sad (Serbia). Her postdoctoral work at the University of Glasgow (the UK) and University of Porto (Portugal) focused on drinking water microbiome in distribution systems and urinary microbiome in women health. Her current interest includes computational approaches in microbial ecology to explore global microbiome and antimicrobial resistome within the EMBARK project. Svetlana is a strong supporter of the open science initiatives, such as the Microbiome Digest, Microbiome Virtual International Forum, National Summer Undergraduate Research Project etc.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
19.10.2022. 18:15h, Matematički fakultet (Online)
Machine Learning for Precision Medicine
Martin Ester, PhD
Video: Recorded lecture (MP4, 85min, 92MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, October 19, starting at 18:15, in an online classroom at the Faculty of Mathematics. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
The vision of precision medicine is to diagnose patients more accurately and treat them more effectively taking into account their individual genomic, life-style, and environmental factors. Machine learning is expected to play a major role in implementing this vision, using the capability of machine learning methods to learn from very complex training datasets and to produce consistent, repeatable predictions (diagnoses, prognoses, treatment recommendations) for test cases. In this talk, I will focus on three of our own works in this area. Machine learning models typically exploit strong correlations between input and output variables but scientists want to discover causal mechanisms. We have explored the task of discovering causal relationships from observational data, employing data mining methods to generate candidate causal relationships and adopting quasi-experimental design to test the significance of the candidates. Our method HUME discovers single causes in the application domain of pharmacogenomics, using network analysis for candidate generation. Biomedical datasets tend to be small and high-dimensional. Fortunately, related public datasets are available, and transfer learning is a promising approach to increase the effective dataset size. We have used drug response prediction as our driving application, where we want to transfer from large pre-clinical datasets to small clinical datasets. We have proposed AITL, which adjusts for the discrepancies in both the input and the output space, employing adversarial domain adaptation and multi-task learning. Machine learning models for precision medicine need to be explainable. Our method DBKANN adopts a knowledge-based approach that employs the available biological knowledge on how proteins form complexes and act together in pathways to form the architecture of a deep neural network. BDKANN does not only achieve high accuracy but also enables meaningful explanations of individual predictions and the discovery of novel connections in the biological network. In the last part of the talk, I will introduce single-cell RNA sequencing, a new technology which creates gene expression profiles at single-cell resolution, and enables new types of deeper biomedical analyses such as the discovery of cell-types and their relevance for phenotype prediction. I will sketch some of our ongoing work in the area of single cell data science.
Lecturer
Martin Ester received a PhD in Computer Science from ETH Zurich, Switzerland, in 1989. He has been working for Swissair developing expert systems before he joined University of Munich as an Assistant Professor in 1993. Since November 2001, he has first been an Associate Professor and now a Full Professor at the School of Computing Science of Simon Fraser University. From May 2010 to April 2015, he has served as the School Director. Dr. Ester has published extensively in the top conferences and journals of his field such as ACM SIGKDD, WWW, ACM RecSys, ISMB, and PSB. According to Google Scholar, his publications have received more than 51'000 citations, and his h-index is 69. He received the KDD 2014 Test of Time Award for his paper on DBSCAN, was elected as a Fellow of the Royal Society of Canada in 2019, and was appointed Distinguished Professor at Simon Fraser University in 2021. Martin Ester’s research interests are in the area of data mining and machine learning, with a current focus on transfer learning, causal discovery and inference, explainable machine learning, and clustering. Many of the driving applications of his research are in the biomedical field, and he has an honorary appointment as Senior Research Scientist at the Vancouver Prostate Center.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
25.05.2022. 18:15h, Matematički fakultet (Onlajn)
Bioinformatics Research Groups at Faculty of Mathematics and Faculty of Biology of University of Belgrade
dr Nenad Mitić, dr Anđela Rodić
A meeting of the Bioinformatics seminar will be held on Wednesday, May 25, starting at 18:15, in an online classroom at the Faculty of Mathematics. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
The closure of the spring season of the Bioinformatics Seminar will be next Wednesday, May 25th. We will have a presentation of two local bioinformatics research groups and their current projects - from Faculty of Mathematics and Faculty of Biology, both University of Belgrade. Our lecturers will be prof. Dr Nenad Mitic (MatF) and Dr Andjela Rodic (BioF). On this occasion, the presentation will be in Serbian.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
11.05.2022. 18:15h, Matematički fakultet (Onlajn)
Prediction of alphabets of local protein structures using data mining methods
Mirjana Maljković, PhD
Video: Recorded lecture (MP4, 52min, 60MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, May 11, starting at 18:15, in an online classroom at the Faculty of Mathematics. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
The 3D structure of the backbone can be described using prototypes of local protein structures. A set of local structure prototypes determines the library of local protein structures, also called the structural alphabet. A structural alphabet is defined as a set of N prototypes of L amino acid length.
Amongst several approaches to the prediction of 3D structures from amino acid sequences, one approach is based on the prediction of SA prototypes for a given amino acid sequence. Protein Blocks (PBs) is the most known SA, and it is composed of 16 prototypes of five consecutive amino acids.
In the research, models for PBs prediction from sequence information were developed using different data mining approaches. The amino acid sequences were combined with the results of the following tools: Spider3 predictor of protein structure properties, several predictors of the protein’s intrinsically disordered regions and a tool for finding repeats in amino acid sequences. Obtained data were used as an input to the prediction model of structural alphabet prototypes. The highest accuracy of the constructed models is 80%. The previous best available prediction has accuracy of 61%. The best-achieved accuracy was for the model constructed using the C5.0 algorithm.
Lecturer
Mirjana Maljković, PhD, is a teaching assistant at the Department of Computer Science and Informatics at the Faculty of Mathematics, University of Belgrade. Mirjana has participated in providing lectures in the following subjects: Relational Databases, Database Programming, Software Development, Data Mining and Introduction to Programming. Her main areas of interest are data mining and bioinformatics. She is a member of BioMath: Bioinformatics research group at the Faculty of Mathematics, University of Belgrade.
Seminar
The organizer of the seminar is BIRBI. The heads of the seminar are Prof. dr Gordana Pavlović-Lažetić, Prof. dr Nataša Pržulj and dr Anđela Rodić.
13.04.2022. 18:15h, Matematički fakultet (Onlajn)
Dve decenije toksikogenomike – mogućnosti i izazovi in silico toksikogenomičke analize podataka
Dr sc. Danijela Đukić-Ćosić
U sredu 13. aprila, sa početkom u 18 sati i 15 minuta, u onlajn učionici na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika. Pozivaju se nastavnici i studenti doktorskih i master studija računarskih, matematičkih, bioloških i drugih srodnih disciplina da iskoriste priliku da nam se pridruže.
Rezime
Tradicionalna ispitivanja toksičnosti na eksperimentalnim životinjama ograničena su vremenom, troškovima i etičkim principima. U novije vreme postoji sve veći interes za razvoj, testiranje i korišćenje novih metoda in silico analize koje primenom računarskih tehnika mogu da pruže efikasniju procenu toksičnosti, a time i rizika po zdravlje ljude, kako pri izloženosti lekovima tako i svakoj drugoj supstanci. Značajan doprinos ovim metodama pruža toksikogenomika, koja kombinacijom tradicionalne toksikologije i bioinformatike sa –omika tehnologijama, genomikom, transkriptomikom, proteomikom i metabolomikom istražuje uticaj izloženosti različitim supstancama na gene, proteine i metabolite u okviru ćelija ili tkiva organizma. Rezultati toksikogenomičkih ispitivanja nastoje da rasvetle molekularne mehanizme toksičnosti i utvrde potencijalne biomarkere koji predviđaju toksičnost ili genetsku preosetljivost. Ovakvi rezultati, prikupljeni i sadržani u bazama podataka poslednjih decenija, i njihovo analiziranje, mogu da posluže kao osnova za razvoj novih naučnih hipoteza. Veruje se da toksikogenomička analiza podataka može biti zamena ili komplementarni pristup za proučavanje bezbednosti budućeg leka tokom pretkliničkih ispitivanja ili pak za predviđanje toksičnosti izazvane sve većim prisustvom različitih supstanci u našem okruženju. Štaviše, toksikogenmička analiza podataka obezbeđuje procenu toksičnih efekata smeše, uzimajući u obzir sve potencijalne interakcije na nivou gena, proteina i metabolita koje mogu biti značajne u ispoljavanju toksičnosti (antagonističko, aditivno/sinergističko dejstvo). Stoga bi se analiza toksikogenomičkih podataka mogla posmatrati kao važna odskočna daska za dalja in vitro i in vivo ispitivanja, omogućavajući smanjenje vremena i troškova celokupnih istraživanja. U ovom izlaganju biće predstavljena toksikogenomička analiza podataka sa sažetim rezimeom potencijalno korisnih softvera i alata, prvenstveno Komparativne toksikogenomičke baze podataka (engl. Comparative Toxicogenomics Database, CTD), uključujući prednosti, ali i ograničenja ovakvih istraživanja.
Predavač
Dr sc. Danijela Đukić-Ćosić je vanredni profesor za užu naučnu oblast toksikologija na Farmaceutskom fakultetu Univerziteta u Beogradu i gostujući profesor na Medicinskom fakultetu Univerziteta u Banjoj Luci na odseku Farmacija. Poseduje kvalifikaciju Evropski registrovani toksikolog, specijalista je farmacije za toksikološku procenu rizika i specijalsita toksikološke hemije. Ima više od dve decenije istraživačkog iskustva na polju toksikologije različitih supstanci, prventsveno metala i endokrinih ometača (ftalati i bisfenol A), mehanizama toksičnosti, procene rizika po zdravlje ljudi i toksikologije smeša, zatim polju istraživanja o mogućnostima da se umanje štetni efekti toksičnih supstanci primenom magnezijuma i probiotskih kultura, a poslednjih godina istraživanja su osim na in vivo modelima usmereni i na in silico istraživanja, odnosno toksikogenomičku analizu podataka za utvrđivanje relacija između izloženosti toksičnim supstancama i razvoj bolesti, predviđenje ciljnih mesta dejstva smeše toksičnih supstanci, neželjenih efekata kombinovane primene lekova itd. Učestvuje u nacionalnim i međunarodnim istraživačkim projektima i projektima u oblasti obrazovanja. Trenutno je angažovana na Erazmus+ projektu “Podsticajna nastava za aktivno učenje i uspeh studenata u onlajn okruženju”/Effective teaching for student engagement and success in digital learning environment – (StudES) (KA226 — Partnerships for Digital Education Readiness), lokalna je kontakt osoba za CEEPUS mrežu — CEEPUS SI-0905 “Training and research in environmental chemistry and toxicology”, istraživač je na PROMIS projektu “Dekodiranje uloge ekspozoma u endokrinom zdravlju” (DecodExpo) (Fond za nauku, Republika Srbija) i rukovodilac trogodišnjeg istraživačko razvojnog projekta međunarodne saradnje sa Narodnom Republikom Kinom “Povećanje efikasnosti imunoterapije karcinoma kombinacijom CAR-T ćelija ili PD-1/PD-L1 inhibitora sa imunomodulatorima” (Ministarstvo prosvete, nauke i tehnološkog razvoja, Republika Srbije).
Seminar
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nataša Pržulj i dr Anđela Rodić.
23.03.2022. 18:15h, Matematički fakultet (Onlajn)
Numerical simulations of microwave liver tumor ablation
Branislav Rađenović i Marija Radmilović-Rađenović
Video: Recorded lecture (MP4, 67min, 84MB)
A meeting of the Bioinformatics seminar will be held on Wednesday, March 23, starting at 18:15, in the online classroom at the Faculty of Mathematics. Teachers and students of doctoral and master studies in computer science, mathematics, biology and other related disciplines are invited to join us.
Abstract
Hepatocellular carcinoma accounts for around 75% of all liver cancers, and represents the fourth most common cause of cancer-related deaths. Microwave ablation is a well esatblished treatment of hepatocellular carcinoma. The success rate for completely eliminating small liver tumors in patients treated with microwave ablation isgreater than 85%. Microwave ablation is also highly recommended for COVID-19 patients with liver tumors as a fast treatment with a short recovery time. The involvement of the temperature dependence of the heat capacity, the thermal conductivity, and blood perfusion, is pivotal for establishing the correct ablation process and preserving the healthy tissue.
Every mathematical model for the simulation of microwave ablation consists of three fundamental components. The first component is the model of the antenna probe (or applicator) that generates a microwave field in the tissue. The antennas are usually mechanically and geometrically complex, and the simulation relies on having accurate electromagnetic material and tissue properties. In this study, we use a compact 10-slot microwave antenna with an impedance pi-matching network that creates near-spherical ablation zones. The second component describes the heat distribution in the tissue including sources and sinks and the phase changes. Heat transfer during the MWA process can be accurately described by the Pennes bioheat equation. In our case, the microwave field is the source of heat, and the heat sinks are represented by the blood perfusion term in the heat transfer equation. The third part deals with the effect of heat on tumor cells and their destruction. All these components of the ablation model depend on a variety of material parameters, which themselves depend on the various states of the tissue. Finally, to define realistic simulation model, we were using the data from the 3D-IRCADb-01 database of hepatocellular carcinoma.
The 3D finite elements method (FEM) is used to solve coupled electromagnetic-field and heat-transfer equations, including all details of antenna design and properties of healthy and tumoral tissue. Our 3D model is created within the COMSOL Multiphysics FEM-based simulation platform.
Lecturers
Branislav Radjenović was born in 1954. He received the B.Eng., M.Sc., and Ph.D. degrees (1990) from the Faculty of Electrical Engineering, University of Belgrade. He joined the Military Technical Institute, Belgrade, in 1980, where he worked on the design of weapon systems and telecommunication devices. Since 1993, he taught courses in solid state electronics, electromagnetics, optoelectronics, television technique and digital electronics at the Military Technical Academy, Belgrade. Since 2008, he has been the principal research fellow at the Institute of Physics Belgrade, University of Belgrade, He also has been a visiting professor at POSTECH, University of Science and Technology in Pohang, S. Korea and at Comenius University, Faculty of Mathematics, Informatics and Physics in Bratislava, Republic of Slovakia. He has been a Supervisor to several Ph.D. and M.Sc. theses. He has won the National Scholarship Programme by Slovak Academic Information (2018). Currently, Dr. Radjenović is a member of the Program IDEAS project-SimSurgery.
Dr. Branislav Radjenović was the author/co-author of more than 140 articles in international scientific journals, 7 chapters in the books, one certified Technical solution, and more than 30 lectures at the international conferences. His research interests include computational physics, level set method, finite element method, biomedical applications, MEMS technologies, etc. Also, he is a principal software developer at the MaSaTECH research and development company in Bratislava, specialized in Ion Mobility Spectrometry devices.
Prof. Dr. Marija Radmilović-Radjenović obtained her B.Sc. degree at the Faculty of Mathematics, University of Belgrade and completed her M.Sc. and Ph.D. degrees at the Faculty of Physics, University of Belgrade. She is the principal research fellow at the Institute of Physics Belgrade and a part-time professor at the Faculty of Physics, University of Belgrade, for the course “Methods of numerical simulations in physics of ionized gases and plasmas”. Dr. Radmilović-Radjenović has been a visiting professor at POSTECH, University of Science and Technology in Pohang, S. Korea and at Comenius University, Faculty of Mathematics, Informatics and Physics in Bratislava. She also spent shorter sabbaticals at Ruhr University in Bochum, Germany and at the Faculty of Informatics and Information Technologies, the Slovak University of Technology in Bratislava, Republic of Slovakia. She has won the annual award of the Institute of Physics Belgrade (2008) and the National Scholarship Programme by Slovak Academic Information twice (2015) and (2020). Marija Radmilović-Radjenović is a member of the Editorial Board of the Open Physics (previously Central European Journal of Physics) and was co-editor of the books “Radicals and Non-Equilibrium Processes in Low-Temperature Plasmas” and “Argon: Production, Characteristics and Applications.” She is a member of the scientific committees of two international conferences-International Conference on Phenomena in Ionized Gases (ICPIG) and Symposium on Applications of Plasma Processes (SAPP), She was a Secretary and a member of the scientific committee of 5th EU-Japan Workshop on Plasma Processing and a member of the Council of Multidisciplinary Studies at the University of Belgrade. Dr. Radmilović-Radjenović is a member of the Center of Excellence in Photonics and was a member of numerous national and international projects. Currently, she is a leader of the Program IDEAS project-SimSurgery.
Dr. Marija Radmilović-Radjenović was author/co-author of more than 150 articles in international scientific journals, 5 chapters in the books, one certified Technical solution, and more than 40 lectures at international conferences. Her research interests include computational physics, level set method, finite element method, biomedical applications, plasma-based technologies, plasma surface interactions, etc.
Seminar
The heads of the seminar are Prof. Dr. Gordana Pavlović-Lažetić, Prof. Dr. Nataša Pržulj i Dr. Anđela Rodić.
02.03.2022. 18:15h, Matematički fakultet (Onlajn)
Infusing Structure and Knowledge Into Biomedical AI
Marinka Zitnik
U sredu 2. marta, sa početkom u 18 sati i 15 minuta, u onlajn učionici na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika. Pozivaju se nastavnici i studenti doktorskih i master studija računarskih, matematičkih, bioloških i drugih srodnih disciplina da iskoriste priliku da nam se pridruže.
Abstract
Grand biomedical challenges require AI/ML models to generalize to entirely new scenarios not seen during training. However, standard supervised learning is incredibly limited in scenarios, such as designing novel therapeutics, modeling emerging pathogens, and treating rare diseases. In this talk, I present our efforts to overcome these obstacles by infusing structure and knowledge into AI/ML algorithms. First, I will outline general-purpose and scalable algorithms for few-shot learning on graphs. At the core is the notion of local subgraphs that transfer knowledge from one task to another, even when only a handful of labeled examples are available. This principle is theoretically justified as we show the evidence for predictions can be found in subgraphs surrounding the targets. Finally, I will conclude with applications in drug discovery and precision medicine. The algorithmic predictions were validated in human cells and led to discovering a new class of drugs.
Lecturer
Marinka Zitnik (https://zitniklab.hms.harvard.edu) is an Assistant Professor at Harvard University with appointments in the Department of Biomedical Informatics, Broad Institute of MIT and Harvard, and Harvard Data Science. She is an internationally recognized expert on graph analytics, studying machine learning and applications to science, medicine, and health. Dr. Zitnik has published extensively in top ML venues (e.g., NeurIPS, ICLR, ICML) and leading interdisciplinary journals (e.g., Nature Methods, Nature Communications, PNAS). She has organized numerous conferences in the nexus of AI, deep learning, drug discovery, and medical AI at leading conferences (NeurIPS, ICLR, ICML, ISMB, AAAI, WWW), where she is also in the organizing committees. She is also an ELLIS Scholar in the European Laboratory for Learning and Intelligent Systems (ELLIS) Society. Her research recently won best paper and research awards from the International Society for Computational Biology, Bayer Early Excellence in Science Award, Amazon Faculty Research Award, Rising Star Award in Electrical Engineering a nd Computer Science (EECS), and Next Generation Recognition in Biomedicine, being the only young scientist with such recognition in both EECS and Biomedicine.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nataša Pržulj i dr Anđela Rodić.
15.12.2021. 18:15h, Matematički fakultet (Onlajn)
Kinetika specifičnih antitela u periodu do 9 meseci posle primene vakcina protiv SARS-CoV-2 dostupnih u Srbiji i njihova efektivnost
dr Olgica Đurković-Đaković
U sredu 15. decembra, sa početkom u 18 sati i 15 minuta, u onlajn učionici na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika. Pozivaju se nastavnici i studenti doktorskih i master studija računarskih, matematičkih, bioloških i drugih srodnih disciplina da iskoriste priliku da nam se pridruže.
Rezime
U Institutu za medicinska istraživanja se sprovodi istraživanje "Praćenje prirodno stečenog i vakcinalnog imuniteta na SARS-CoV-2 virus kod odraslih ispitanika", čiji je osnovni cilj utvrđivanje dugoročne evolucije antitela specifičnih za SARS-CoV-2 kod pacijenata sa COVID-19 i kod vakcinisanih ispitanika.
Posle prvog preseka rezultata ovog istraživanja predstavljenih na ovom istom forumu krajem marta ove godine, u okviru ovog predavanja biće predstavljeni rezultati dobijeni praćenjem naše kohorte ispitanika u periodu do 9 meseci posle vakcinacije.
Predavač
Dr sc. med. Olgica Đurković-Đaković je naučni savetnik Instituta za medicinska istraživanja Univerziteta u Beogradu. Osnivač je Centra izuzetnih vrednosti za zoonoze prenošene hranom i vektorima, kao i Nacionalne referentne laboratorije za toksoplazmozu. Ima višedecenijsko iskustvo u istraživanjima u oblasti infekcija gde je rukovodila nizom naučnoistraživačkih projekata. Od početka pandemije aktivno se sa svojim Centrom uključila u istraživanja dijagnostičkih, imunoloških i epidemioloških aspekata kroz longitudinalni monitoring SARS-CoV-2 specifičnih antitela u različitim kategorijama populacije.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nataša Pržulj i dr Anđela Rodić.
01.12.2021. 18:15h, Matematički fakultet (Onlajn)
Reducing Imprecision in Precision Medicine
Igor Jurisica, PhD, DrSc
U sredu 1. decembra, sa početkom u 18 sati i 15 minuta, u onlajn učionici na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika. Pozivaju se nastavnici i studenti doktorskih i master studija računarskih, matematičkih, bioloških i drugih srodnih disciplina da iskoriste priliku da nam se pridruže.
Abstract
To fathom complex health and disease processes, we need to systematically integrate diverse types of information, including multiple high-throughput datasets and diverse annotations. However, thousands of potentially important proteins remain poorly characterized. Models are going to be only as good as the networks used to build them are comprehensive. Computational biology methods, including machine learning and data mining can help fill these gaps with accurate predictions, making disease modeling more comprehensive. In turn, they help to find hidden signal in data often rejected as noise.
Such models and integrative analyses help us develop new hypotheses, answer complex questions such as what factors cause disease; which patients are at high risk; will patients respond to a given treatment; how to rationally select a combination therapy to individual patient, etc. In turn, this enables data-driven precision medicine. However, AI-based prediction systems can have great cohort-based accuracy, yet fail on specific patient level. Thus, subgroup performance and individual prediction confidence must be assessed and understood in combination with model explanation, before translating AI-based prediction tools into clinical practice.
Lecturer
Igor Jurisica, PhD, DrSc is a Senior Scientist at Osteoarthritis Research Program, Division of Orthopedic Surgery, Schroeder Arthritis Institute and Krembil Research Institute, Professor at University of Toronto and Visiting Scientist at IBM CAS. He is also an Adjunct Professor at the Department of Pathology and Molecular Medicine at Queen's University, and an adjunct scientist at the Institute of Neuroimmunology, Slovak Academy of Sciences. Since 2015, he has also served as Chief Scientist at the Creative Destruction Lab, Rotman School of Management, and since 2021 he is a scientific director of the World Community Grid.
His research focuses on integrative informatics and the representation, analysis and visualization of high-dimensional data to identify prognostic/predictive signatures, determine clinically relevant combination therapies, and develop accurate models of drug mechanism of action and disease-altered signaling cascades. He has published extensively on data mining, visualization and integrative computational biology, including multiple papers in Science, Nature, Nature Medicine, Nature Methods, J Clinical Oncology, J Clinical Investigations. He has been included in Thomson Reuters 2014, 2015 & 2016 lists of Highly Cited Researchers (http://highlycited.com), and The World's Most Influential Scientific Minds: 2015 & 2014 Reports. In 2019, he has been included in the Top 100 AI Leaders in Drug Discovery and Advanced Healthcare list (Deep Knowledge Analytics, http://analytics.dkv.global).
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nataša Pržulj i dr Anđela Rodić.
21.04.2021. 18:15h, Matematički fakultet,
Računarske metode particionisanja i grupisanja u biološkim mrežama
Milana Grbić
Video: Recorded lecture (MP4, 62min, 57MB)
U sredu 21. aprila, sa početkom u 18 sati, u onlajn učionici na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika. Pozivaju se nastavnici i studenti doktorskih i master studija računarskih, matematičkih, bioloških i drugih srodnih disciplina da iskoriste priliku da nam se pridruže.
Abstract
U doktorskoj disertaciji "Računarske metode particionisanja i grupisanja u biološkim mrežama" razmatrano je nekoliko problema definisanih nad biološkim mrežama, koji su rješavani optimizacionim metodama i metodama mašinskog učenja. Rješavani su sljedeći problemi: particionisanje rijetkih bioloških mreža na k-plex podmreže (engl. Maximum Edge-weight k-plex Partitioning Problem - Max-EkPP), predviđanje uloge metabolita u metaboličkim reakcijama, particionisanje bioloških mreža na visoko povezane komponente (engl. Highly Connected Deletion problem – HCD) i identifikacija značajnih grupa proteina dodavanjem novih grana u težinsku mrežu proteinskih interakcija (engl. protein-protein interaction – PPI). Računarski najzahtjevnija faza posljednjeg razmatranog problema uključuje problem određivanja minimalnog broja dodatnih PPI, koje treba uključiti u PPI mrežu, da bi poznati proteinski kompleksi, posmatrani kao podgrafovi u toj mreži, bili povezani. Za razmatrane probleme su analizirani i razvijani odgovarajući matematički modeli, koji su dalje rješavani različitim računarskim metodama. Pored rješavanja navedenih problema sa računarskog aspekta, u disertaciji se istražuje dalja primjena dobijenih rezultata u oblastima biologije i biohemije, kao i integracija rezultata unutar postojećih bioinformatičkih alata.
Predavač
Doc. dr Milana Grbić je nastavnik na Studijskom programu za matematiku i informatiku, Prirodno-matematičkog fakulteta Univerziteta u Banjoj Luci (www1, www2).
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
07.04.2021. 18:15h, Matematički fakultet,
Bioinformatics analysis of correlation between protein function and intrinsic disorder
Goran Vinterhalter
U sredu 07. aprila, sa početkom u 18 sati, u onlajn učionici na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika. Pozivaju se nastavnici i studenti doktorskih i master studija računarskih, matematičkih, bioloških i drugih srodnih disciplina da iskoriste priliku da nam se pridruže.
Abstract
The correlation of molecular function and protein intrinsic disorder is an important aspect of understanding the relationship between function, sequence and structure. This research was inspired by statistical correlation evaluation method described by Xie et al. (J Proteome Res 6 (2007) 1882–1898, reference study), where the authors analyzed the relationship between structure and function of proteins from Swiss-Prot database and where these functions were described with Swiss-Prot function keywords. In this research, we investigated whether the conclusions from the reference study stand for another dataset with richer functional annotation. We used CAFA3 challenge training dataset where the function was described with terms from Gene Ontology (GO terms). In order to compare the results with the previous work, we associated the GO terms with the corresponding Swiss-Prot function keywords. The results were compared with the reference study by first repeating the analysis with Swiss-Prot function keywords and then by GO terms. We used PONDR VSL2b disorder predictor to label over 66,000 CAFA3 proteins as putatively disordered or ordered. Out of 186 Swiss-Prot keywords (belonging to molecular function type) with more than 20 annotated proteins, we found 47 to be highly order related and 44 highly disorder related. Using the same dataset and annotation constraints, out of 1781 GO term (belonging to molecular function type), we found 746 to be highly order related and 564 highly disorder related. Comparison of two functional annotations, GO and Swiss-Prot keywords, showed consistent results in cases when it was possible to map a Swiss-Prot keyword to a corresponding GO term. Because of the small number of such cases, we propose a new method for deriving the missing mappings between Swiss-Prot keywords and GO terms with the highest likelihood by measuring similarity (Jaccard index) between sets of protein annotated by different functions. Comparison with results from the reference study revealed prevalence of binding related functions (disorder related) in the current dataset even though the same functions were not present in previous results.
Predavač
Goran Vinterhalter je master informatičar Matematičkog fakulteta, trenutno zaposlen u kompaniji Agilent Technologies u Gentu, Belgija
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
24.03.2021. 18:15h, Matematički fakultet, (Onlajn),
Monitoring prirodno stečenog u odnosu na vakcinalni imunitet SARS-CoV-2 – prvi presek
dr Olgica Đurković-Đaković
U sredu 24. marta, sa početkom u 18 sati, u onlajn učionici na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika. Pozivaju se nastavnici i studenti doktorskih i master studija računarskih, matematičkih, bioloških i drugih srodnih disciplina da iskoriste priliku da nam se pridruže.
Apstrakt
U Institutu za medicinska istraživanja se sprovodi istraživanje "Praćenje prirodno stečenog i vakcinalnog imuniteta na SARS-CoV-2 virus kod odraslih ispitanika", čiji je osnovni cilj utvrđivanje dugoročne evolucije antitela specifičnih za SARS-CoV-2 kod pacijenata sa COVID-19 i kod vakcinisanih ispitanika. Potonja populacija je posebno značajna u svetlu činjenice da zbog široke dostupnosti četiri različite vakcine Srbija predstavlja jedinstveni poligon za komparativnu analizu vakcinalnog imuniteta. Osim doprinosa globalnom razumevanju specifičnog humoralnog imunskog odgovora u simptomatskim i asimptomatskim slučajevima COVID-19, kao i kod osoba vakcinisanih protiv SARS-CoV-2 virusa, očekuje se da podaci o nivou imunološke zaštite postignute prirodnom i veštačkom imunizacijom doprinesu daljem planiranju ciljane prevencije npr identifikacijom kategorija populacije koje je potrebno češće revakcinisati, te da budu osnova za izradu prediktivnih modela daljeg toka SARS-CoV-2 epidemije u ovom delu Evrope i šire. Na ovom predavanju će biti predstavljen prvi presek rezultata ovog istraživanja.
Predavač
Dr sc. med. Olgica Đurković-Đaković je naučni savetnik Instituta za medicinska istraživanja Univerziteta u Beogradu. Rukovodilac je Centra izuzetnih vrednosti za zoonoze prenošene hranom i vektorima, kao i Nacionalne referentne laboratorije za toksoplazmozu. Rukovodila je nizom naučnoistraživačkih projekata od kojih jednim iz FP6 programa Evropske komisije, i učestovala u više međunarodnih projekata i drugim oblicima međunarodne saradnje. Angažovana je kao nastavnik na doktorskim studijama na Medicinskom fakultetu UB. Posebno je ponosna na svoju ulogu u obrazovanju mladih, uključujući mentorstvo doktorskih disertacija (8 odbranjenih i 2 u izradi) i magistarskih teza (3 odbranjene). Ukupni indeks naučne kompetentnosti iznosi > 650 (Scopus: radova 117, citiranost: > 1500, H-indeks 22). Bavi se različitim aspektima infekcija apikompleksnim patogenima, uključujući imunopatogenezu, dijagnostiku i eksperimentalnu hemioterapiju. Od početka pandemije aktivno se sa svojim Centrom uključila u istraživanja dijagnostičkih, imunoloških i epidemioloških aspekata kroz longitudinalni monitoring SARS-CoV-2 specifičnih antitela u različitim kategorijama populacije.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
24.02.2021. 18:15h, Matematički fakultet, (Onlajn)
Biofizički i bioinformatički pristup u istraživanju prenošenja virusa SARS-CoV-2 u populaciji
Anđela Rodić
Video: Recorded lecture (MP4, 67min, 73MB)
U sredu 24. februara, sa početkom u 18 sati, u onlajn učionici na Matematičkom fakultetu bice održan sastanak seminara Bioinformatika. Pozivaju se nastavnici i studenti doktorskih i master studija racunarskih, matematickih, bioloških i drugih srodnih disciplina da iskoriste priliku da nam se pridruže.
Apstrakt
Biofizičari sa Biološkog fakulteta su, u saradnji sa naučnicima i nastavnicima sa Instituta za fiziku i Medicinskog fakulteta, sproveli istraživanje uticaja demografskih i klimatskih faktora na prenošenje virusa-izazivača bolesti COVID-19. Istraživanje kombinuje biofiziku (primena nelinearnih dinamičkih kompartmentalnih modela), bioinformatiku (prikupljanje i analiza velike količine podataka) i analizu široko rasprostranjenih obrazaca rasta infekcije ("scaling relations" u biofizici). Dobijeni rezultati ukazuju na to da nekoliko demografskih i meteoroloških faktora značajno utiče na inherentnu prenosivost (osnovni reprodukcioni broj) virusa u populaciji. Takođe je analizirana i disproporcija između intenzivnog širenja infekcije u Vuhanu (Hubei) i mnogo manjih brojeva zaraženih u drugim kineskim provincijama. Predloženo je da se ova zagonetka može objasniti kombinacijom značajno veće inherentne prenosivosti virusa u Vuhanu (R0 koje zavisi upravo od faktora sredine) i veće efikasnosti mera za suzbijanje epidemije u drugim provincijama. Ukupno gledano, rezultati ovih analiza ukazuju na to da dinamika širenja epidemije može značajno da zavisi od potencijalno visoko heterogenih i naizgled slučajnih faktora, kao što su varijacije u demografskim i meteorološkim uslovima, kao i od njihove složene interakcije sa uvedenim merama kontrole. Razumevanje ovih faktora je ključno, ne samo za analizu rizika tokom pandemije, već i za dugoročno predviđanje ponašanja virusa u populaciji u slučaju da bolest postane endemična.
Predavač
Anđela Rodić je asistent Biološkog fakulteta na Katedri za opštu fiziologiju i biofiziku. Kao član grupe prof. Marka Đorđevića, Anđela se bavi istraživanjima u oblasti sistemske biologije, odnosno biofizičkog modelovanja bioloških sistema.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
10.02.2021. 18:15h, Matematički fakultet, (Onlajn)
Between viral targets and differentially expressed genes in COVID-19: the sweet spot for therapeutic intervention
Nataša Pržulj
U sredu 10. februara, sa pocetkom u 18 sati, u onlajn učionici na Matematičkom fakultetu bice održan prvi ovogodišnji sastanak seminara Bioinformatika. Pozivaju se nastavnici i studenti doktorskih i master studija racunarskih, matematickih, bioloških i drugih srodnih disciplina da iskoriste priliku da nam se pridruže.
Abstract
The COVID-19 pandemic is raging. It revealed the importance of rapid scientific advancement towards understanding and treating new diseases. To address this challenge, we build onto our previous methods for extracting new biomedical knowledge from the wiring patterns of systems-level, heterogeneous biomedical networks. These methods are needed due to the flood of molecular and clinical data, measuring interactions between various bio-molecules in and around a cell that form large, complex systems. These systems-level network data provide heterogeneous, but complementary information about cells, tissues and diseases. The challenge is how to mine them collectively to answer fundamental biological and medical questions. This is nontrivial, because of computational intractability of many underlying problems on networks (also called graphs), necessitating the development of approximate algorithms (heuristic methods) for finding approximate solutions.
We adapt an explainable artificial intelligence algorithm for data fusion and utilize it on new omics data on viral-host interactions, human protein interactions, and drugs to better understand SARS-CoV-2 infection mechanisms and predict new drug-target interactions for COVID-19. We discover that in the human interactome, the human proteins targeted by SARS-CoV-2 proteins and the genes that are differentially expressed after the infection have common neighbors central in the interactome that may be key to the disease mechanisms. We uncover 185 new drug-target interactions targeting 49 of these key genes and suggest re-purposing of 149 FDA-approved drugs, including drugs targeting VEGF and nitric oxide signaling, whose pathways coincide with the observed COVID-19 symptoms. Our integrative methodology is universal and can enable insight into this and other serious diseases.
Predavač
Professor Natasa Przulj is an ICREA Research Professor and a Group Leader at Barcelona Supercomputing Center. She is a leader in network science and AI algorithms for biomedical data fusion applied to precision medicine. Her research has been cited around 10,000 times, h-index=43, i10-index=70 (Google Scholar) and supported by over €15 million in competitive funding. Notably, she received three prestigious, single PI, European Research Council (ERC) grants: Consolidator (2018-2023), Proof of Concept (2020-2022) and Starting (2012-2017). She has been elected into: The Serbian Royal Academy of Scientists and Artists in 2019; Academia Europaea, The Academy of Europe, in 2017; Fellow of the British Computer Society (BCS) Academy of Computing, in 2013. In 2014, she received a BCS Roger Needham Award, sponsored by Microsoft Research, in recognition of the potential her research has to revolutionize health and pharmaceutics. She obtained a PhD in Computer Science from the University of Toronto.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
16. 10. 2019. 18:15h, Matematički fakultet, sala 718
Are Proteins composed of little structural bricks?
Alexandre G. de Brevern
U sredu 16. oktobra, sa početkom u 18:15 sati, u sali 718 na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Abstract
The protein structures are classically described as composed of two regular states, the α-helices and the β-strands and one non-regular and variable state, the coil. Nonetheless, the representation of only two repetitive states hides other interesting repetitive structures, e.g., PolyProline II helix [1], or super secondary motifs. The definition of secondary structures is often considered as fixed and ideal. In fact, the rules for secondary structure assignments are complex and can also bias our analyses [2]. It is so interesting to look at other complementary description such as sets of small prototypes or "structural alphabets", able to analyze local protein structures and to approximate every part of the protein backbone. The principle of a structural alphabet is simple. A set of average local protein structures is firstly designed. They approximate (efficiently) every part of the structures. As one residue is associated to one of these prototypes, we can translate the 3D information of the protein structures as a series of prototypes (letters) in 1D, as the amino acid sequence.
Structural alphabets have also been used to predict the protein backbone conformation and in ab initio / de novo methods. Our structural alphabet is composed of 16 mean protein fragments of 5 residues in length, called Protein Blocks (PBs) [3]. They have been used both to describe the 3D protein backbones and to perform a local structure prediction. PBs have been cited in more than 350 publications worldwide from the prediction of long fragments, to definition of binding site [4]. We have used this approach to compare / superimpose protein structures. The assessment of a simple approach done on the classical benchmark sets was surprisingly excellent. It is equivalent or better than the best actual approaches [5] and still is. Moreover, PBs can be used to assess protein flexibility with efficiency [6].
Reference
[1] Mansiaux Y., Joseph A.P., Gelly J.-C., de Brevern A.G. (2011) Assignment of PolyProline II conformation and analysis of sequence - structure relationship Plos One 6(3): e18401.
[2] Tyagi M., Bornot A., Offmann B., de Brevern A.G. (2009) Analysis of loop boundaries using different local structure assignment methods, Protein Science 18(9):1869-81.
[3] de Brevern A.G., Etchebest C. & Hazout S. (2000) Bayesian probabilistic approach for predicting backbone structures in terms of protein blocks, Proteins, 41:271-87.
[4] Joseph A.P., Agarwal G., Mahajan S., Gelly J.-C., Swapna L.S., Offmann B., Cadet F., Bornot A., Tyagi M., Valadié H., Schneider B., Etchebest C., Srinivasan N., de Brevern A.G. (2010) A short survey on Protein Blocks, Biophysical Reviews 2(3):137-145.
[5] Joseph A.P., Srinivasan N., de Brevern A.G. (2012) Progressive structure-based alignment of homologous proteins: Adopting sequence comparison strategies, Biochimie 94:2025-34.
[6] Narwani T.J., Craveur P., Shinada N.K., Floch A., Santuz H., Vattekatte A.M., Srinivasan N., Rebehmed J., Gelly J.-C., Etchebest C., de Brevern A.G. (2019) Discrete analyses of protein dynamics, JBSD (2019) in press.
Predavači
Alexandre G. de Brevern, (1) INSERM UMR-S 1134, DSIMB, (2) Univ Paris, (3) Institut National de Transfusion Sanguine, (4) GR-Ex laboratoire d’excellence, Paris, France.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
22. maj 2019. u 18:15h, Matematički fakultet, sala 718
Dinamika regulacije restrikciono-modifikacionih sistema bakterija: od biofizičkih modela ka sintetičkoj biologiji
Anđela Rodić i Stefan Graovac
U sredu 22. maja, sa početkom u 18:15 sati, u sali 718 na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Apstrakt
Restrikciono-modifikacioni (R-M) sistemi su male genske mreže bakterija, često u potpunosti kodirane plazmidima, koje obezbeđuju ekspresiju dva enzima: restrikciona endonukleaza (RE) seče specifične DNK sekvence, dok ih metiltransferaza (MT) štiti od sečenja. Osnovni zadatak R-M sistema je da brane bakterijsku ćeliju od unete strane DNK, a da pritom ne oštete genom domaćina. Kako bi ekspresija navedenih enzima bila strogo kontrolisana u novonaseljenoj bakteriji i, naročito, da bi uključivala kašnjenje pojave RE koje obezbeđuje dovoljno vremena za početnu, zaštitnu metilaciju genoma, iznenađujuće, R-M sistemi se oslanjaju na veliki broj različitih regulatornih svojstava, kao što su specijalizovani transkripcioni faktori (C proteini), njihovo kooperativno vezivanje, preklopljeni promotori, antisens RNK, odsustvo mesta za vezivanje ribozoma sa transkripta, i dr. Imajući u vidu imunsku funkciju koju dele svi R-M sistemi, pretpostavili smo da nekoliko odabranih osobina dinamike ekspresije sistema, koje su odgovorne za bezbedno i efikasno uspostavljanje sistema u novom domaćinu, uslovljava dizajn regulacije sistema. Zatim smo, kombinacijom termodinamičkog i dinamičkog modelovanja i biohemijskih eksperimenata, analizirali četiri R-M sistema koja koriste veoma različite mehanizme regulacije transkripcije i pokazali da se dizajn svakog od ovih sistema može objasniti predloženim dinamičkim principima [1-4]. Međutim, bakterije u prirodnoj sredini mogu da se nađu u veoma različitim globalnim fiziološkim uslovima, koji vode značajnim razlikama u brzini rasta ćelija. Zato se postavlja pitanje, u kojoj meri efekati dinamike populacije utiču na unutarćelijsku dinamiku ekspresije RE i MT, kao i da li su nađeni dinamički principi robusni u odnosu na promene brzine rasta ćelije. Koristeći prva raspoloživa merenja na nivou pojedinačnih ćelija kod R-M sistema [1], pokazali smo da je uključivanje efekata dinamike populacije neophodno da bi se kvantitativno objasnila eksperimentalna merenja [5]. Takođe, naši preliminarni rezultati ukazuju na to da je regulacija R-M sistema dizajnirana (odnosno optimizovana evolucijom) tako da poveća robusnost dinamičkih osobina sistema (npr. odnosa RE i MT u stacinarnom stanju) u odnosu na promene brzine rasta ćelija. Sa stanovišta bioinžinjerskih primena, raznovrsna regulatorna svojstva i evolucioni principi dizajna pronađeni u R-M sistemima mogu da posluže kao gradivni blokovi, odnosno smernice u konstruisanju veštačkih genskih kola u sintetičkoj biologiji.
Reference
[1] Morozova N, Sabantsev A, Bogdanova E, Fedorova Y, Maikova A, Vedyaykin A, Rodic A, Djordjevic M, Khodorkovskii M, Severinov, Nucleic Acids Research 44:790, 2016.
[2] Rodic A, Blagojevic B, Zdobnov E, Djordjevic M and Djordjevic M, BMC Systems Biology 11:377, 2017.
[3] Rodic A, Blagojevic B and Djordjevic M, In Systems Biology (pp. 37-58). SpringerNature, 2018.
[4] Klimuk E, Bogdanova E, Nagornykh M, Rodic A, Djordjevic M, Medvedeva S, Pavlova O, Severinov K, Nucleic Acids Research 46:10810, 2018.
[5] Graovac S, Rodic A, Djordjevic M, Severinov K, Djordjevic M, Molecules 24:198, 2019.
Predavači
Anđela Rodić i Stefan Graovac, Biološki fakultet.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
8. maj 2019. u 18:15h, Matematički fakultet, sala 718
Kvantifikacija slučajnosti u biološkim kompleksnim mrežama
dr Marija Mitrović-Dankulov
U sredu 8. maja, sa početkom u 18:15 sati, u sali 718 na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Apstrakt
Biological systems can be represented as complex networks, where network nodes represent units of the system, while links represent interactions between them. These networks are neither of regular or random structure, but rather an intricate combination of order and disorder. Scientists have developed a large set of different topological measures for characterization and description of different structural properties of real networks. It turns out that these statistical measures are not independent, i.e., many properties appear as a statistical consequence of a relatively small number of fixed topological properties in a real network. We explore this dependence in two different biological networks, protein-protein interaction and brain network, using the method of dk-series. We find that many important local and global topological properties of protein-protein interaction network are closely reproduced by dk-random graphs whose degree distributions, degree-degree correlations, and clustering are the same as in original real network, while this is only in part true for human brain network. These differences are a consequence of different spacial constraints present during the evolution of these brain networks.
Predavač
Dr Marija Mitrović-Dankulov, Institut za fiziku.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
24. april 2019. u 18:15h, Matematički fakultet, sala 718
Etičko pitanje eutanazije i transplantacije sa medicinskog i socijalnog gledišta
dr sc. med Vladan Čokić
U sredu 24. aprila, sa početkom u 18:15 sati, u sali 718 na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Apstrakt
Tokom 2018. godine doneta su dva nova zakona - ZAKON O LjUDSKIM ĆELIJAMA I TKIVIMA ("Sl. glasnik RS", br. 57/2018) I ZAKON O PRESAĐIVANjU LjUDSKIH ORGANA ("Sl. glasnik RS", br. 57/2018), a trenutno je na uvidu javnosti GRAĐANSKI ZAKONIK REPUBLIKE SRBIJE koji reguliše i "Pravo na dostojanstvenu smrt (eutanaziju)". Tematika je aktuelna i od javnog značaja, a prošla je (tendenciozno ili ne) prilično ispod radara javnosti. ZAKON O LjUDSKIM ĆELIJAMA I TKIVIMA, u članu25, obuhvata i prikupljanje matičnih ćelija iz pupčanika novorođenčeta, gde se u članu 20 navodi da će se sprovesti „sva odgovarajuća medicinska ispitivanja i zahvati u cilju procene i smanjenja fizičkih i psihičkih rizika za zdravlje davaoca“. Da li je to tačno i kakva su iskustva kod nas i u drugim zemljama? Takođe se navodi izuzeće u članu 24: „ne primenjuju se ako se utvrdi da njihovo uzimanje predstavlja minimalni rizik i minimalno opterećenje za davaoca.“ Da li su ovi rizici adekvatno sagledani i predstavljeni potencijalnim davaocima? Koje su prednosti i nedostaci transplantacije matičnih ćelija iz pupčanika novorođenčeta u odnosu na klasične transplantacije matičnim ćelijama iz kostne srži i periferne krvi odraslih osoba? Da li ima i kvalitativne koristi ili samo kvantitativne (povećanje broja raspoloživih donora)? Novi ZAKON O PRESAĐIVANjU LjUDSKIH ORGANA uveo je pretpostavljenu donaciju organa kao odgovor na nedovoljan odaziv za donacije u prethodnom periodu, oslanjajući se na rešenja drugih zemalja iz Evrope po tom pitanju. U Srbiji je oko 2000 ljudi na listi čekanja za transplantaciju, dok na transplantaciju bubrega čeka oko 700 bolesnika, a na dijalizi je oko 6.000 ljudi. Transplantacije se u Srbiji vrše nakon moždane smrti, a u Evropi i nakon cirkulatorne smrti (srčanog zastoja). Kolika je uspešnost transplantacije organa i koje su stope preživljavanja za pojedine organe? Da li su dovoljni klinički dokazi smrti za prekid života potencijalnog donora? Koji su potencijalni budući davaoci organa za transplantaciju? Da li smo odškrinuli vrata ka tome da ćemo svi biti potencijalni davaoci organa bez saglasnosti? Na osnovu člana 92 nacrta GRAĐANSKOG ZAKONIKA dato je pravo na eutanaziju, uz ogradu da će se o tome odlučiti nakon javne rasprave. Eutanazija je praksa namernog okončanja života kako bi se ublažio bol i patnja ili sprečio gubitak dostojanstva, dok se pod terminom „potpomognuto samoubistvo“ podrazumeva da lekar obezbeđuje sredstva da pacijent izvrši samoubistvo. I u jednom i u drugom slučaju lekar se pretvara u direktnog ili indirektnog izvršitelja, što je u suprotnosti sa Hipokratovom zakletvom. Koji su argumenti za eutanaziju, a koji su protiv eutanazije? Eutanazija je odobrena u Holandiji i Belgiji još 2002 i 2003 godine, kakva su njihova iskustva i posledice po društvo? Da li je moguća zloupotreba zakona o eutanaziji i da li zakon može da iskontroliše rizike eutanazije? Zašto su u jednoj temi transplantacije i eutanazija? Zato što eutanazija pruža nove donore za transplantaciju. Mnogo kontroverznih pitanja zahteva šire sagledavanje problematike što dovodi do konkretnih odgovora.
Predavač
Dr sc. med Vladan Čokić je naučni savetnik na Institutu za medicinska istraživanja. Rukovodilac je Grupe za molekularnu onkologiju tog instituta kao i nacionalnog projekta "Ispitivanje patogeneze hematoloških maligniteta" Ministarstva prosvete, nauke i tehnološkog razvoja. Predsednik je Etičkog odbora Instituta za medicinska istaživanja.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
11. april 2019. u 18:15h, Matematički fakultet, sala 718
Primena ciljanog bisulfitnog sekvenciranja u epigenetici kancera
dr Miljana Tanić
U četvrtak 11. aprila, sa početkom u 18:15 sati, u sali 718 na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Apstrakt
Uz par izuzetaka, svaka ćelija multicelularnih eukariota ima identičan genom. Ono što determiniše ćelijski fenotip je kombinacija epigenetičkih markera koje zajedno određuju gde, kada i koliko će se eksprimirati informacija kodirana u genomu. DNK metilacija je jedan od najvažnijih epigenetičkih mehanizama neophodan za ćelijsku diferencijaciju. Izmenjen profil DNK metilacije je rani događaj u evoluciji karcinoma, pri čemu dolazi do lokalne hipermetilacije tumor supresora i globalne hipometlacije.
Metode sekvenciranja nove generacije (NGS) omogućile su proučavanje epigenoma. Za analizu DNK metilacije zlatni standard je bisulfitno sekveniranje (BS-seq), koje pored rezolucije na nivou jednog nukleotida omogućava i faznu rekonstrukciju epialela na nivou jedne očitane sekvence (sequencing read). Analiza celog epigenoma je i dalje izuzetno skupa i bioinformatički zahtevna metoda, te su razvijene metode za usmeravanje sekvenciranja na ciljane regione širom genoma (CpG ostrva, regulatorne sekvence, i druge).
Na seminaru biće predstavljena uporedna analiza dostupnih metoda za ciljano bisulfitno sekvenciranje, opisani tehnički izazovi i specifična bioinformatička obrada ovih podataka. Takođe, biće prikazana primena metoda ciljanog BS-seq na primeru studije intra-tumorske heterogenosti i evolucije nesitnoćelijskog karcinoma pluća (TRACERx studija).
Predavač
Dr Miljana Tanić je naučni saradnik na Institutu za onkologiju i radiologiju.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
27. mart 2019. u 18:15h, Matematički fakultet, sala 718
Istraživanje porekla i srodstva stanovništva naseljenog u Šumadiji tokom seoba u XVIII veku kombinovanjem arhivske građe i DNK analize Y hromozoma
dr sc. med Vladan Čokić
U sredu 27. marta, sa početkom u 18:15 sati, u sali 718 na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Apstrakt
Naseljavanje prostora između planina Kosmaj i Rudnik u XVIII veku uglavnom potiče od dinarske struje, pre svega iz Sjenice i Pešteri. To naseljavanje rezultat je Prve, a naročito Druge seobe Srba 1739 godine. Proređeno stanovništvo je održavalo kontakte iz prapostojbine rodbinskim i kumovskim vezama na širokom prostoru sa retkim naseljima. U napuštena naselja je dolazilo od nekoliko do ne više od desetak porodica. To su starosedeoci Šumadije i iz svake te porodice nastalo je po 10-20 familija kao velika bratstva. Danas se te stare rođačke veze retko pamte ili su deo nesigurnog predanja.
Korišćenjem arhivske građe (deftera, popisa stanovništva, otomanskih i austrijskih poreskih knjiga, crkvenih matičnih knjiga) i DNK analize muškog Y hromozoma obrađeno je 10 bratstava sa oko 150 familija i 300 godina starim korenima na prostoru između planina Kosmaja i Rudnika. Horizontalnim ispitivanjem DNK utvrđene su haplogrupe karakteristične za određena bratstva i srodstvo unutar njih sa zajedničkim poreklom. Vertikalnim ispitivanjem DNK utvrđena je kako stopa mutacije unutar 300 godina starih rodoslovnih stabala tako i nastajanje i migracije haplogrupe do hiljadama godina starog zajedničkog pretka. Ispitivanje zastupljenosti haplogrupa karakterističnih za stanovništvo Šumadije je rađeno na osnovu rodoslovnih stabala i broja familija unutar bratstva, a ne uobičajenim metodama slučajnog uzorka sa geografskim odrednicama. Na ovaj način izbegnuto je preklapanje uzoraka iz istog bratstva sa različitim patronimijama (prezimenima).
Utvrđene haplogrupe u Šumadiji su zatim poređene sa haplogrupama u Srbiji, Balkanu i ostalim slovenskim zemljama. Predstavljena studija je pokazala široku srodnost na ispitivanom području sa zanemarivanjem rođačkih veza po ženskoj liniji i tolerisanjem starih rođakih veza po muškoj liniji.
Predavač
Dr sc. med Vladan Čokić je naučni savetnik na Institutu za medicinska istraživanja. Rukovodilac je Grupe za molekularnu onkologiju tog instituta kao i nacionalnog projekta "Ispitivanje patogeneze hematoloških maligniteta" Ministarstva prosvete, nauke i tehnološkog razvoja.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
16. januar 2019. u 18:15h, Matematički fakultet, sala 718
Integrative bioinformatics with Galaxy - building frameworks to serve the 21th century data science problems
dr. Björn Grüning, Univerzitet u Frajburgu, Nemačka
U sredu 16. januara, sa početkom u 18:15 sati, u sali 718 na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Kratka biografija predavača
Dr Grüning je rukovodilac Freiburg Galaxy projekta i „driving force” u Galaxy zajednici. Doktorirao je u oblasti farmaceutske bioinformatike i radi istraživanja u oblasti genomike i epigenetike (citati). Više o projektu Galaxy možete pogledati na stranici http://www.bioinf.uni-freiburg.de/Galaxy/.
Organizatori
Rukovodioci seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Nenad Mitić i Anđela Rodić.
21. decembar. u 18h, Matematički fakultet, sala 706
CRISPR/Cas i restrikciono-modifikacioni sistemi: modelovanje dinamike ekspresije imunskih sistema bakterija
Anđela Rodić i Bojana Blagojević
U sredu 21. decembra, sa početkom u 18 sati, u sali 706 na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Apstrakt
Restrikciono-modifikacioni (RM) sistemi bakterija tipično kodiraju dva enzima: 1) restrikcioni enzim (R), koji prepoznaje specifične ciljne sekvence na DNK i seče ih, i 2) DNK metilazu (M) koja, metilovanjem istih sekvenci, sprečava dejstvo R na njih. RM sistemi putem plazmida mogu dospeti u nove bakterijske domaćine. Od suštinskog je značaja da RM sistemi poseduju mehanizme za koordinisanu ekspresiju svojih gena, koji će osigurati da se R sintetiše sa zakašnjenjem u odnosu na M, dovoljnim da do početka njegovog delovanja genom domaćina bude zaštićen metilovanjem. Za regulaciju ekspresije R i M je u nekim RM sistemima zadužen transkripcioni faktor S, takođe kodiran sistemom. Nosilac RM sistema postaje rezistentan na infekciju virusom čija DNK sadrži nemetilovane ciljne sekvence.
RM sistemi se mogu odlikovati različitim tipovima organizacije gena (konvergentnom i divergentnom) i regulatornim mehanizmima (uloga konstante dimerizacije S, kooperativnosti u vezivanju proteina za DNK, brzine translacije, i dr.). Kako bismo sistematski ispitali efekte ovih regulatornih svojstava na odlike dinamike ponašanja sistema, razvili smo i perturbisali biofizičke modele regulacije ekspresije gena za jedan konvergentan i jedan divergentan RM sistem. Zaključili smo da se različite osobine konstrukcije i regulacije kombinuju u RM sistemima tako da obezbede nekoliko univerzalnih odlika dinamike ekspresije gena, koje su zadate odbrambenom funkcijom RM sistema.
Odbrambenu funkciju u prokariotima ima i napredni imunski sistem CRISPR/Cas koji može da prepozna i uništi stranu DNK komplementarnu malim crRNK, koje nastaju sečenjem transkripta CRISPR niza (pre-crRNK) pomoću Cas proteina. U fiziološkim uslovima, CRISPR i cas promotori su utišani globalnim represorima koji se kooperativno vezuju za DNK, a mehanizam kojim se sistem indukuje prilikom infekcije je nepoznat. Na osnovu sličnosti CRISPR/Cas sa RM sistemima u mehanizmima regulacije transkripcije, kao i funkcionalnim dinamičkim zahtevima, ispitali smo značaj kooperativne regulacije u indukciji CRISPR/Cas sistema tako što smo, u biofizičkom modelu, indukovali promotore ovog sistema uvođenjem transkripcione kontrole dobro izučenog RM sistema. Ovim putem smo ukazali na značajnu ulogu kooperativne regulacije u obezbeđivanju veoma brzog prelaska sistema iz isključenog u uključeno stanje.
Predavači
Anđela Rodić (Biološki fakultet, Univerzitet u Beogradu) i Bojana Blagojević (Institut za fiziku, Univerzitet u Beogradu)
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
7. decembar. u 18h, Matematički fakultet, sala 706
RNK sekvenciranje i transkriptom biljaka: specifičnosti, alati i problemi
Dragana Dudić
U sredu 7. decembra, sa početkom u 18 sati, u sali 706 na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Apstrakt
RNK sekvenciranje (RNA-seq) je procedura sekvenciranja koja omogućava analizu transkriptoma i ekspresije gena, a koja je omogućena pojavom tehnika nove generacije sekvenciranja (NGS). RNA-seq se bitno razlikuje od DNK sekvenciranja u nekoliko aspekata. Sam proces kreiranja sekvenci je specifičan jer se, zbog nestabilnosti RNK molekula, reverzibilnom transkripcijom generišu cDNK biblioteke koje se ubacuju u NGS sekvencioner i sekvenciraju. Dalje, analiza podataka obuhvata, između ostalog, i mapiranje na referentne sekvence. S obzirom da su, čak i za najistraživanije organizme, transkriptomi i dalje nekompletni, ovo mapiranje se u stvari vrši na genom. Zbog toga uobičajeni alati, poput BLAST-a ili ClustalW, nisu pogodni za upotrebu, već se razvijaju novi koji uzimaju u obzir specifičnost RNK podataka. Dodatni problemi nastaju u situacijama kada se analiziraju poliploidni ili visoko repetitivni transkriptomi, kakvi su npr. transkriptomi biljaka. U okviru izlaganja biće dat osvrt na ove teme i predstavljeni konkretni izazovi sa kojima su se istraživači suočili na primeru transkriptoma kukuruza.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
23. novembar. u 18h, Matematički fakultet, sala 706
Human Brain Project - Quo vadis?
prof. dr Pavle Anđus
U sredu 23. novembra, sa početkom u 18 sati, u sali 706 na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Apstrakt
Human Brain Project (HBP) pokrenut 2013. godine prodat je Evropskoj Komisiji od strane harizmatičnog neurobiologa Henrija Markrama (Švajcarski Federalni Tehnološki Institut u Lozani, EPFL) kao hrabar, novi put ka razumevanju mozga, lečenju neuroloških bolesti i razvoju informacionih tehnologija. HBP je jedan od dva H2020 „flagship“ projekta (drugi se bavi grafenom) „teških“ po 1 G€. Cilj projekta je ambiciozno postavljen kao „sticanje fundamentalnih znanja o tome šta znači biti čovek, kako razviti nove tretmane bolesti mozga i izgraditi revolucionarne nove informacione i komunikacione tehnologije“. Smatra se da danas po prvi put moderna IKT ove ciljeve čini dostupnim. Projekat je planirano da čini 10 godina interdisciplinarne saradnje preko 400 naučnika iz 112 institucija u 24 evropske zemlje. Mreža učesnika HBP je otvorenog tipa i organizovan je u 12 subprojekata koji pokrivaju 6 informatičkih platformi, organizaciju mozga, kognitivnu neuronauku, teoriju, etiku i društvo i menadžment.
U predavanju uz originalne HPB slajdove biće reči o razvoju ovog projekta i o njegovom statusu danas. Biće otvorena pitanja o mestu koje u ovom projektu zauzima neurofiziolgija mozga odnosno informacione tehnologije i neuroinformatika.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
9. novembar. u 18h, Matematički fakultet, sala BIM
Baza podataka eksperimentalno utvrđenih neuređenih regiona proteina
prof. dr Nenad Mitić
U sredu 9. novembra, sa početkom u 18 sati, u sali BIM na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Apstrakt
Proteini koji nemaju potpuno uređenu 3D strukturu imaju značajnu ulogu u različitim vrstama ćelijskih procesa. U poslednje dve decenije eksperimentalno je potvrđeno postojanje velikog broja proteina čiji pojedini regioni imaju neuređenu strukturu. Podaci o ovakvim proteinima su prikupljeni i organizovani u bazu podataka proteina sa neuređenim regionima. Prva verzija baze je objavljena 2005. godine u SAD, i ažurirana je sve do 2013. godine. Posle tri godine pauze, u Evropi je formirana nova ažurirana i proširena baza sa povećanim brojem karakteristika proteina.
Na predavanju će biti izložene karakteristike stare i nove verzije baze, prikaz raspoloživih resursa, kao i (kritički) osvrt na postojeće programe za predviđanje neuređenih regiona proteina, uzimajući raspoložive podatke iz baze kao osnovu.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
6. april. u 18h, Matematički fakultet, sala BIM
Viral: Realne simulacije konkurentnih procesa nad multipleks mrežama
Petar Veličković i Andrej Ivašković
U sredu 6. aprila, sa pocetkom u 18 sati, u sali BIM na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Apstrakt
Precizno modelovanje procesa koji se šire predstavlja ključni izazov savremene bioinformatike, naročito u kontekstu predviđanja posledica epidemija (na primer, proporcije stanovništva koja će biti zahvaćena nekom zarazom u kritičnom trenutku). Veliki je broj postojećih pristupa; nedavna otkrića u polju multipleks mreža dozvoljavaju integraciju nekoliko konkurentnih procesa i direktnije modelovanje njihovih interakcija. Međutim, istraživanja su se do sada najviše vršila na nasumično generisanim mrežama i pretpostavke u vezi sa njihovom dinamikom ne moraju neophodno odgovarati ponašanju ljudi. Kao značajan korak ka pravljenju kontrolisanih eksperimenata ove vrste, predstavljamo Viral, sistem zasnovan na multipleks mrežama za simulacije konkurentnih procesa epidemija i informisanosti u savremenom društvu, u kontekstu stvarnog sveta, zasnovan na nezahtevnoj distribuiranoj Android aplikaciji i centralnog servera predviđenog za simulaciju, pri čemu su obe komponente jednostavne za pokretanje i podešavanja. Obezbeđena je infrastruktura za detaljno beleženje i analiziranje rezultata simulacije.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
11. decembar 2015. u 18h, Matematički fakultet, sala BIM
Data driven research in (cancer) genomics
prof. dr Cenk Sahinalp
U petak 11. decembra, sa početkom u 18 sati, u sali BIM u Računarskoj laboratoriji na Matematičkom fakultetu bice održan sastanak seminara Bioinformatika.
Tema sastanka je Data driven research in (cancer) genomics. Predavanje će održati prof. dr Cenk Sahinalp iz Centra za bioinformatička istraživanja Fakulteta za informatiku i računarstvo u Blumingtonu, u Indijani, SAD. Oblast naučnog interesovanja prof. Sahinalpa uključuje bioinformatiku, teorijsko računarstvo i algoritme. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Sequencing projects involving thousands of individual genomes are underway and the need for algorithmic speedup is bigger than ever. We will go through some of the algorithmic developments introduced by my lab to address challenges in big data genomics, especially for cancer research. These algorithms involve one or more techniques in data compression, streaming, memory hierarchy awareness and parallelization. Application areas range from read mapping to variant calling, novel isoform and fusion gene identification to clonality inference.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
2. decembar 2015. u 18h, Matematički fakultet, sala BIM
Prepoznavanje proteinskih ponovaka i ispitivanje odnosa struktura-funkcija proteina uključenih u patogenezu malignih bolesti
dr Nevena Veljković, dr Vladimir Perović
U sredu 2. decembra, sa početkom u 18 sati, u sali BIM u Računarskoj laboratoriji na Matematičkom fakultetu biće održan sastanak seminara Bioinformatika.
Tema sastanka je Prepoznavanje proteinskih ponovaka i ispitivanje odnosa struktura-funkcija proteina uključenih u patogenezu malignih bolesti. Predavanje će održati dr Nevena Veljković i dr Vladimir Perović iz Centra za multidisciplinarna istraživanja Instituta za nuklearne nauke Vinča. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Većina proteina sadrži konzervirane i funkcionalno značajne periodičnosti koje ne odgovaraju konkretnoj aminokiselinskoj sekvenci, već su asocirane sa određenim fizičkohemijskim karakteristikama aminokiselina. Ove neočigledne periodičnosti u proteinskoj sekvenci mogu se detektovati kompjuterskim alatima koji se zasnivaju modelovanju i različitim oblicima reprezentacije sekvence. Metode zasnovane na EIIP parametru i deskriptorima hidrofobnosti i Furijeovoj transformaciji razvijane u Institutu za nuklearne nauke Vinča pokazale su se veoma efikasne za prepoznavanje funkcionalno značajnih mutacija i interaktivnih domena proteina. Cilj ovog predavanja je da predstavi naša najnovija istraživanja i projekte za buduće saradnike i doktorante.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
18. novembar 2015. u 18h, Matematički fakultet, sala BIM
Modelovanje imunih sistema bakterija
dr Anđela Rodić
U sredu 18. novembra, sa početkom u 18 sati, u sali BIM u Računarskoj laboratoriji na Matematičkom fakultetu bice održan sastanak seminara Bioinformatika.
Tema sastanka je Modelovanje imunih sistema bakterija. Predavanje će održati dr Anđela Rodić sa Biološkog fakulteta. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
U okviru predavanja će biti predstavljen uvod u modelovanje regulacije i dinamike ekspresije gena. Kao primer iz tekućeg istraživanja biće izloženi rezultati modelovanja tri "imuna" sistema kod bakterija. Ovi sistemi predstavljaju primer strategije ekspresije molekula koji su s jedne strane korisni, budući da ćeliju brane od virusa, a s druge strane toksični zbog mogućeg autoimunog odgovora. Konkretno, modelovaće se restrikciono-modifikacioni (RM) sisteme AhdI i Esp1396I kod kojih je par otrov/antiotrov predstavljen restrikcionom endonukleazom i metilazom, kao i nedavno otkriveni, napredni imuni sistem kod prokariota CRISPR/Cas. Biće predstavljeni i najnoviji rezultati koji se odnose na modelovanje dinamike ekspresije Esp1396I sistema gde su dobijena teorijska predviđanja upoređena sa prvim raspoloživim eksperimentalnim podacima na nivou pojedinačnih ćelija.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
21. oktobar 2015. u 18h, Matematički fakultet, sala BIM
Primena tehnologije "Next Generation Sequencing (NGS)" u medicinskoj genetici
dr Maja Tarailo-Graovac
U sredu 21. oktobra, sa početkom u 18 sati, u sali BIM u Računarskoj laboratoriji na Matematičkom fakultetu bice održan sastanak seminara Bioinformatika.
Tema sastanka je Primena tehnologije "Next Generation Sequencing (NGS)" u medicinskoj genetici. Predavanje će održati dr Maja Tarailo-Graovac. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
2003. god je završen projekat otkrivanja DNK sekvence ljudskog genoma. Taj projekat je koštao skoro 3 biliona dolara i trajao je više od 10 godina. Danas jedan ljudski genom (gdje je svaka baza pokrivena makar 30 puta) se rutinski sekvencira za cenu nižu od 1500 dolara. Napredak u tehnologiji sekvenciranja kao i računarskoj obradi podataka je od ključnog značaja za Medicinsku genetiku i Medicinu danas. Tokom seminara, dr Maja Tarailo-Graovac će da iznese podatke svoga rada na obradi NGS sekvenci kod model organizma Caenorhabditis elegans kao i obradi sekvenciranih ljudskih egzoma i genoma kod pacijenata oboljelih od rijetkih genetskih bolesti.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
27. maj 2015. u 18h, Matematički fakultet, sala 718
Modelling the Hypothalamic-Pituitary-Adrenal (HPA) axis as a complex oscillating reaction system
dr Ljiljana Kolar Anić
U sredu 27. maja , sa početkom u 18 sati, u sali 718 na Matematičkom fakultetu bice održan sastanak seminara Bioinformatika.
Tema sastanka je Modelling the Hypothalamic-Pituitary-Adrenal (HPA) axis as a complex oscillating reaction system. Predavanje će održati dr Ljiljana Kolar Anić, redovni profesor Fakulteta za fizičku hemiju Univerziteta u Beogradu. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
To analyze dynamic states in a reaction system, the system’s evolution in time is described using a set of differential equations. These equations can be obtained either by postulating a stoichiometric model or by a more general mathematical approach, using differential equations with nonlinear terms to mimic the temporal behaviour of the investigated system. Even though it is not possible to assign a mechanism with complete certainty to any real chemical reaction, the stoichiometric approach to modelling is advantageous because it is consistent with known biochemical data, model predictions can be more directly compared with real experiments and the model can be more easily expanded to account for additional species that modulate the main process.
To demonstrate the validity of the stoichiometric approach for describing complex biological systems, we shall consider a low-dimensional model of the Hypothalamic-Pituitary-Adrenal (HPA) axis, which is a highly dynamical structure that integrates and controls the functions of the nervous and endocrines systems under normal physiological conditions and stress [1-5]. We also show that the model can be easily expanded to account for additional reactions and additional species, such as cholesterol, which is the only precursor of steroid hormones and modulates the HPA axis dynamics. [6]
Reference
- [1] Jelić S, Čupić Ž, Kolar-Anić Lj. Mathematical modeling of the hypothalamic-pituitary-adrenal system activity. Math Biosci. 2005, 197:173-187.
- [2] Jelić S, Čupić Ž, Kolar-Anić Lj, Vukojević V. Predictive Modelling of the Hypothalamic-Pituitary-Adrenal (HPA) function. Dynamic Systems Theory Approach by Stoichiometric Network Analysis and Quenching Small Amplitude Oscillations. Int J Nonlin Sci Num, 2009, 10:1451-1472.
- [3] Marković V. M, Čupić Ž, Vukojević V, Kolar-Anić Lj. Predictive modeling of the hypothalamic-pituitary-adrenal (HPA) axis response to acute and chronic stress. Endocr J. 2011, 58:889-904.
- [4] Marković VM, Čupić Ž, Ivanović A, Kolar-Anić Lj. The stability of the extended model of hypothalamic-pituitary-adrenal (HPA) axis examined by stoichiometric network analysis (SNA). Russ. J. Phys. Chem. A, 2011 85:2327-2335.
- [5] Čupić Ž, Marković V, Ivanović A, Kolar-Anić Lj. Modeling of the Complex Nonlinear Processes: Determination of the Instability Region by the Stoichiometric Network Analysis, In: Christopher R. Brennan, Ed. Mathematical Modelling, Nova Science Publishers Inc., New York, 2011, pp. 111-178, ISBN: 978-1-61209-651-3.
- [6] Marković, V.M, Čupić, Ž, Maćešić, S, Stanojević, A, Vukojević, V, Kolar-Anić, Lj, Modelling cholesterol effects on the dynamics of the hypothalamic–pituitary–adrenal (HPA) axis, Mathematical Medicine and Biology (2014), doi:10.1093/imammb/dqu020.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
13. maj 2015. u 18h, Matematički fakultet, sala 718
Toksoplazma i Shizofrenija: molekularna veza
dr Aleksandra Uzelac
U sredu 13. maja , sa početkom u 18 sati, u sali 718 na Matematičkom fakultetu bice održan sastanak seminara Bioinformatika.
Tema sastanka je Toksoplazma i Shizofrenija: molekularna veza. Predavanje će održati dr Aleksandra Uzelac, iz Instituta za medicinska istraživanja, Beograd. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Toksoplazmoza je parazitska zoonoza izazvana protozoom Toxoplasma gondii iz roda Apicomplexa. Smatra se da je T. gondii jedan od najuspešnijih intracelularnih parazita, s obzirom na svoju ubikvitarnu distribuciju i činjenicu da inficira gotovo trećinu čovečanstva. Kod više od 80% inficiranih imunokompetentnih osoba toksoplazmoza protiče asimptomatski, dok kod imunokompromitovanih osoba može izazvati teške kliničke posledice sa dramatičnom simptomatologijom i mogućim smrtnim ishodom.
Kada je u metabolički aktivnoj formi (tahizoit), parazit može da inficira i da se razmnožava u svakoj nukleisanoj ćeliji domaćina (akutna infekcija) a da potom, uslovljen aktiviranim mehanizmima imunske odbrane domaćina, snizi svoje metaboličke funkcije i konvertuje se u latentnu formu (bradizoit) koju karakteriše formiranje intracelularnih cista. Predilekciona tkiva za formiranje cista T. gondii su mozak, retina, srčani i skeletni mišići. Bradizoiti opstaju u dormantnom incistiranom obliku praktično doživotno jer se ciste ne mogu ukloniti ni imunskim odgovorom domaćina niti antiparazitarnim lekovima. Zapravo, posebna karakteristika ovog parazita, koja ga je istorijski i dovela u vezu sa neuropsihijatrijskim oboljenjima, je njegova sposobnost da uspešno pređe krvno-moždanu barijeru i formira ciste u neuronima i glija ćelijama svog domaćina.
Eksperimenti na glodarima su pokazali da infekcija T. gondii može da ima dramatičan uticaj na ponašanje životinja, dok je uticaj infekcije na ponašanje ljudi još uvek predmet brojnih kontroverzi. Pokazalo se ipak u nekim kliničkim studijama da je prevalenca antitela specifičnih za T. gondii kod pacijenata obolelih od shizofrenije nešto viša u odnosu na opštu populaciju, što je uslovilo razvoj istraživanja na molekularnom nivou o T. gondii kao mogućem uzročniku ili faktoru rizika za razvoj shizofrenije.
U sklopu našeg istraživanja, nastojimo da kroz bioinformatički pristup na molekularnom nivou identifikujemo vrstu i stepen povezanosti infekcije T. gondii sa različitim aspektima shizofrenije.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
15. april 2015. u 18h, Matematički fakultet, sala 718
Nelinearna dinamika mikrotubula
dr Slobodan Zdravković
U sredu 15. aprila, sa početkom u 18 sati, u sali 718 na Matematičkom fakultetu bice održan sastanak seminara Bioinformatika.
Tema sastanka je Nelinearna dinamika mikrotubula. Predavač će biti dr Slobodan Zdravković, iz Instituta za nuklearne nauke Vinča. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Mikrotubule su važan dio ćelijskog kostura. To su dugački šuplji valjci unutrašnjeg prečnika oko 15 a spoljašnjeg oko 25 nanometara. Prostiru se od jedra do ćelijske membrane. Red veličine dužine mikrotubule je od mikrometra do milimetra. Ove dugačke postoje u nervnim ćelijama.
U predavanju će biti pokazana njihova građa. Mikrotubule se sastoje od dugačkih struktura koje čine omotač pomenutog valjka. Zovu se protofilamenti i najčešće ih je 13. Sastoje se od dimera koji su, za modele koji će biti objašiynjeni u predavanju, osnovna jedinica građe. Bitno je da je dimer električni dipol. Biće objašnjene interakcije između dimera i izvor nelinearnosti. Vidjećemo kako se nelinearna dinamika mikrotubula opisuje matematičkim aparatom i dolazi do nelinearnih diskretnih i kontinualnih diferencijalnih jednačina. Rješenja dobijena do sada su kink-solitoni, lokalizovani modulisani talasi zvani brideri i, odnedavno, zvonasti solitoni. Ta rješenja su dobijena i analitički i numerički a u predavanju će biti objašnjeni samo neki analitički postupci.
Pored mehaničke uloge mikrotubule igraju važnu ulogu u ćelijskoj diobi a predstavljaju i put po kome se kreću motor proteini. Za očekivati je da je ovim bjelančevinama potrebna pobuda da bi se pokrenule. Može se očekivati da su to spomenuti talasi. Naime, hidrolizom adenozin trifosfata se oslobodi energija koja neki dimer, ili grupu dimera, pobudi na oscilovanje. Ta pobuda se prostire duž mikrotubule i kada stigne do motor proteina za njega predstavlja okidač.
Na kraju će biti prodiskutovane mogućnosti eksperimentalnih istraživanja.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
1. april 2015. u 18h, Matematički fakultet, sala 718
COST akcije i značaj povezivanja i saradnje u oblasti bioinformatike
dr Vesna Pajić
U sredu 1. aprila, sa početkom u 18 sati, u sali 718 na Matematičkom fakultetu bice održan sastanak seminara Bioinformatika.
Tema sastanka su COST akcije i značaj povezivanja i saradnje u oblasti bioinformatike. Predavač će biti dr Vesna Pajić, rukovodilac Centra za istraživanje podataka i bioinformatiku Poljoprivrednog fakulteta Univerziteta u Beogradu. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Discipline razvoja i primene matematičkih, statističkih i računarskih tehnika na istraživanja podataka u biološkim naukama, jednim imenom nazvane bioinformatika, imaju izrazito interdisciplinaran karakter i zahtevaju dobro poznavanje i matematičke teorije koja leži u osnovi metoda koje se koriste, kao i biologije i različitih bioloških fenomena koji se istražuju. To čini obrazovanje naučnog i stručnog kadra iz oblasti bioinformatike veoma složenim procesom, koji je u Srbiji tek u povoju. Sa druge strane, potrebe za ovim kadrom rastu iz dana u dan. Jedan od načina podizanja istraživačkih kapaciteta u oblasti bioinformatike, sem formalnog sistema obrazovanja, jeste i uključivanje u aktivnosti koje se sprovode u okviru COST akcija sa odgovarajućim temama.
COST akcije su mehanizam pomoću koga EU omogućava i podržava međunarodnu naučnu saradnju u različitim oblastima. Njihova dinamika i način funkcionisanja čine ih jednim od najjednostavnijih instrumenata ovakvog tipa. Glavne aktivnosti COST akcija su radni sastanci, treninzi i kratkotrajne naučne posete. Prijavljivanje i učešće u aktivnostima je relativno jednostavno, a same aktivnosti su fokusirane na određene teme, koje svako može sam da odabere i uskladi sa svojim oblastima istraživanja.
U okviru ovog izlaganja biće prvo predstavljene COST akcije uopšte, a zatim i dve COST akcije iz oblasti NGS (Next Generation Sequencing) tehnika, obrade podataka i primena. Prva od njih “Next Generation Sequencing Data Analysis Network” je upravo završena, ali je izuzetno značajna zato što se bavila NGS tehnologijom sa stanovišta matematike i računarstva, te je kao svoje rezultate najavila izdavanje jednog sveobuhvatnog priručnika za upotrebu NGS softvera i metoda. Takođe, ova akcije se bavila načinom edukacije bioinformatičara, pa je i u tom smislu dala veoma korisne smernice na koji način je efikasno da se ta edukacija sprovodi. Posebno, veliki broj materijala sa različitih konferencija i radionica je i dalje dostupan sa sajta ove akcije. Druga akcija, pod nazivom “Application of next generation sequencing for the study and diagnosis of plant viral diseases in agriculture”, je tek započela u martu 2015. Ova akcija stavlja akcenat na primenu NGS tehnologije u poljoprivredi, pa su i njene aktivnosti podeljene u, sa jedne strane, bioinformatičke aktivnosti koje razvijaju samu tehnologiju, i, sa druge strane, aktivnosti koje se bave primenama u biološkim naukama. Osim predstavljanja samih akcija, u okviru predavanja biće predočeni i konkretni primeri kako učestvovati u COST akcijama.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
Prezentacija
Odavde možete preuzeti prezentaciju: (PPS).
18. mart 2015, 18h, Matematički fakultet, sala 718
Biofizička analiza transkripcione regulacije kod bakterija i bakteriofaga
Jelena Guzina
U sredu 18. marta, sa pocetkom u 18 sati, u sali 718 na Matematičkom fakultetu bice održan sastanak seminara Bioinformatika.
Tema sastanka je Biofizička analiza transkripcione regulacije kod bakterija i bakteriofaga. Predavač će biti Jelena Guzina, doktorant na Biološkom fakultetu. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Transkripcija predstavlja početni korak genske ekspresije, a kod prokariota i glavnu tačku njene regulacije. Vezivanjem holoenzima polimeraze RNK za regulatorne sekvence uzvodno od gena – promotora, započinje njena inicijacija.
Od mogućnosti precizne detekcije promotora, glavnih signala transkripcione inicijacije, uveliko zavisi razumevanje njene regulacije, kao i predviđanje gena. Ovo je značajan bioinformatički izazov, budući da standardne metode za detekciju regulatornih elemenata vode ka velikom broju lažnih pozitiva. Osnovni uzrok niske preciznosti, pri primeni standardnih metoda za predviđanje motiva na detekciju promotora, je značajna varijabilnost sekvenci koje definišu kanonski σ70 promotor. Zato je neophodno kvantitativno razumevanje parametara koji opisuju funkcionalni promotor, što je ispitivano kroz primenu biofizičkog modela transkripcione inicijacije u kombinaciji sa unapređenim bioinformatičkim metodima.
Organizatori
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
4. mart 2015, 18h, Matematički fakultet, sala 718
Bioinformatika na Matematičkom fakultetu
Nenad Mitić
U sredu 4. marta, sa pocetkom u 18 sati, u sali 718 na Matematičkom fakultetu bice održan prvi ovogodišnji sastanak seminara Bioinformatika. Pozivaju se nastavnici i studenti doktorskih i master studija racunarskih, matematickih, bioloških i drugih srodnih disciplina da iskoriste priliku da se upoznaju sa aktivnostima koje se u ovoj veoma dinamicnoj i atraktivnoj oblasti odvijaju u našoj sredini.
Tema prvog sastanka je Bioinformatika na Matematičkom fakultetu. Predavač na uvodnom sastanku će biti prof. dr Nenad Mitić.
Organizatori seminara su prof. dr Gordana Pavlović-Lažetić, prof. dr Branko Dragović i prof. dr Nenad Mitić, rukovodilac.
12. decembar 2012, 18h, Matematički fakultet, sala 718
Klasterovanje podataka o ekspresiji gena – Pristup baziran na ponovo upotrebljivim komponentama
Milan Vukićević
U sredu, 12. decembra, sa početkom u 18 sati, u sali 718 na Matematičkom fakultetu Univerziteta u Beogradu, u okviru seminara Bioinformatika će biti održano predavanje Milana Vukićevića sa temom Klasterovanje podataka o ekspresiji gena – Pristup baziran na ponovo upotrebljivim komponentama. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Klasterovanje podataka o ekspresiji gena igra važnu ulogu u biomedicinskim istraživanjima. Primena i evaluacija algoritama klasterovanja je sve češća i sve se više ističe važnost primene ovih algoritama u različitim oblastima biomedicine. Podaci o ekspresiji gena predstavljeni u vidu mikronizova, koji se mogu klasterovati na dva načina. Sa jedne strane, geni se mogu klasterovati na osnovu bioloških uzoraka, a sa druge strane biološki uzorci se mogu klasterovati na osnovu ekspresije gena. Veći problem je klasterovanje bioloških uzoraka na osnovu ekspresije gena, gde mnogi algoritmi ne mogu da identifikuju lokalne strukture zbog visoke dimenzionalnosti podataka i malog uzorka. Iako postoje neke preporuke za izbor algoritama za klasterovanje bioloških podataka, ne postoji konsenzus o najboljem algoritmu za tako težak zadatak klasterovanja. Za rešavanje ovog problema, pristup ponovo upotrebljivih komponenti za algoritme klasterovanja bi mogao da da jasniji pravac.
Predavač
Milan Vukićević je asistent na Fakultetu organizacionih nauka, Univerziteta u Beogradu, u Centru za poslovno odlučivanje. Trenutno je na završnoj godini doktorskih studija. Njegovi glavni pravci istraživanja su: dizajn algoritama klasterovanja i klasifikacije, meta-učenje i primena otkrivanja zakonitosti u podacima (eng. data mining) u bioinformatici i edukaciji.
31. oktobar 2012, 18h, Matematički fakultet, sala 718
Jedan osvrt na metode simulacije bioloških molekula
Ognjen Perišić
U sredu, 31. oktobra, sa početkom u 18 sati, u sali 718 na Matematičkom fakultetu Univerziteta u Beogradu, u okviru seminara Bioinformatika će biti održano predavanje dr Ognjena Perišića sa temom Jedan osvrt na metode simulacije bioloških molekula. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Predavanje će predstaviti pregled tehnika korišćenih za simuliranje bioloških molekula u njihovom prirodnom okruženju. Problem vizualizacije bioloških molekula i analiziranje njihovog ponašanja biće uvod u kome će se ukratko pokriti osnove molekularne biologije, kao što su central dogma of molecular biology (DNA-> RNA -> protein), koncept ključa i brave (lock and key concept), i potreba za razumevanjem dinamike biopolimera. Nakon toga pokriće se klasična molekularna dinamika bazirana na korišćenju koncepta polja sila (classical force field) i molekularna dinamika bazirana na principima kvantne mehanika. Treći deo predavanja biće posvećen simplifikovanim metodama koje "žrtvuju" tačnost zarad efikasnosti. Te metode koriste se kada se simuliraju veliki molekularni kompleksi koje je nemoguće efikasno interpretirati korišćenjem klasičnih metoda zbog njihove numeričke zahtevnosti.
Predavač
Ognjen Perišić je rođen u Beogradu, diplomirao na ETF (BSc), doktorirao na University of Illinois at Chicago (Department of bioengineering, program in bioinformatics). Između diplomiranja na ETF-u i postdiplomskih studija bavio se programiranjem. Oblasti istraživanja su mu proteinska dinamika i statistička fizika.
June 26-28th, 2012, University of Belgrade - Faculty of Mathematics
International Meeting on Data Mining in Bioinformatics
We are pleased to announce that International Meeting on Data Mining in Bioinformatics will take place in Belgrade, Serbia, on June 26-28th, 2012, at the University of Belgrade - Faculty of Mathematics.
Scope
The purpose of the International Meeting on Data Mining in Bioinformatics is to bring together researchers interested in the development and application of data mining and information technologies to the field of life sciences. The focus of this meeting is on data mining. However, areas of interest include other information technologies, as well, such as sequence analysis, biostatistics, pattern recognition, machine learning and other related fields.
Another purpose of the meeting is promoting bioinformatics among researches and prospective researches here in Belgrade and Serbia, broadening and deepening our own understanding of the field and bringing together different research communities, from informatics to molecular biology to biomedicine, thus helping us consider possibilities for future cooperation with colleagues from different fields.
Meeting Site
For more information, please visit the meeting site: bioinfo.matf.bg.ac.rs/dmbi2012.
The tentative programme is available here.
Registration is free and required. Please register.
16. maj 2012, 18h, Matematički fakultet, sala 718
Nova metoda analize pridruživanja – primena na AA indeksa i genotipske / fenotipske karakteristike prokariota
Jasmina Dragoljević
U sredu, 16. maja, sa početkom u 18 sati, u sali 718 na Matematičkom fakultetu Univerziteta u Beogradu, u okviru seminara Bioinformatika će biti održano predavanje Jasmine Dragoljević sa temom Nova metoda analize pridruživanja – primena na AA indeksa i genotipske / fenotipske karakteristike prokariota. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Pronalaženje veza i odnosa parova promenljivih u velikim skupovima podataka je izuzetno važno, jer ti odnosi omogućavaju bolji uvid u različite delove koji zajedno čine celinu, a postiže se analizom tih podataka. Grupa autora je u radu Detecting Novel Associations in Large Data Sets, publikovanom u časopisu Science, prezentovala novu meru zavisnosti između dve promenljive koja je nazvana maksimalni koeficijent informacije (MIC). MIC određuje kako funkcionalne tako i nefunkcionalne veze među podacima. Ona se može koristiti kod skupa podataka sa velikim brojem promenljivih koje poseduju važne i neotkrivene veze. Ova mera pripada grupi statistika koje se zajedno nazivaju MINE i mogu se koristiti ne samo da identifikuju veze između promenljivih, već i za njihovu karakterizaciju po osobinama, kao što su nelinearnost i monotonost. Ona poseduje dve značajne osobine: uopštenost i pravednost, kao i mnoge karakteristike. Dobre mere smanjuju potrošnju vremena i prostora u istraživanju podataka i namenjene su za izbor i rankiranje obrazaca. MINE je koristan, kako za identifikovanje podataka, tako i što otvara mogućnost ispitivanja svih potencijalno zanimljivih odnosa u podacima, nezavisno od njihovog oblika. On nudi nov kriterijum pretraživanja podataka.
Ovo izlaganje ima za cilj prikaz MINE mera kao i aplikacije MINE, razvijene na osnovu optimizovanog algoritma za računanje ovih mera i njenu primenu na AA indekse i genotipske/fenotipske karakteristike prokariota.
Predavač
Jasmina Dragoljević je student doktorskih studija računarstva na Matematičkom fakultetu Univerziteta u Beogradu.
28. mart 2012, 18h, Matematički fakultet, sala 718
Radni sastanak
U sredu, 28. marta, sa početkom u 18 sati, u sali 718 na Matematičkom fakultetu Univerziteta u Beogradu, u okviru seminara Bioinformatika će biti održan radni sastanak. Na sastanku će biti predstavljeni planovi za naredni period, sa posebnim osvrtom na aktivnosti na organizaciji konferencije "Data mining in Bioinformatics" koju Grupa za bioinformatku Matematičkog fakulteta organizuje krajem juna ove godine.
Pozivamo sve zainteresovane da prisustvuju sastanku.
7. mart 2012, 18h, Matematički fakultet, sala 718
Oscilatorni procesi
dr Ljiljana Kolar-Anić
U sredu, 7. marta, sa početkom u 18 sati, u sali 718 na Matematičkom fakultetu Univerziteta u Beogradu, u okviru seminara Bioinformatika će biti održano predavanje dr Ljiljane Kolar-Anić sa temom Oscilatorni procesi. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Oscilatorni procesi su deo našeg života i predmet su izučavanja mnogih nauka. Cilj ovog predavanja je da, u najkraćim crtama, na konkretnim primerima, poveže bazična znanja iz različitih naučnih oblasti neophodna za objašnjenje ovog fenomena, njegovog modeliranja i eventualne primene.
Predavač
Dr Ljiljana Kolar-Anić je redovni profesor Fakulteta za fizičku hemiju Univerziteta u Beogradu. Predaje Statističku termodinamiku i Dinamiku nelinearnih procesa. Na doktorskim studijama zadužena je za nastavu iz predmeta: Oscilatorni procesi u hemijskim, fizičkohemijskim i biološkim sistemima, Modeliranje i simulacija složenih procesa, kao i Biofizička hemija i samoorganizacija neravnotežnih sistema. Bila je rukovodilac više projekata iz oblasti Nelinearnih nauka. Trenutno je rukovodilac nacionalnog Projekta 172015: Dinamika nelinearnih fizičkohemijskih i bioloških sisitema sa modeliranjem i predviđanjem njihovih ponašanja pod neravnotežnim uslovima. Osnovni interes njenog naučnog rada su: Statistička termodinamika, Neravnotežna termodinamika, Hemijska kinetika, Dinamika i samoorganizacija neravnotežnih procesa, Stohastička analiza procesa, Nukleacija, Adsorpcija.
25. januar 2012, 18h, Matematicki fakultet, sala 718
Sekvenciranje genoma i transkriptoma
dr Ana Simonovic, naucni saradnik
U sredu, 25. januara, sa pocetkom u 18 sati, u sali 718 na Matematickom fakultetu Univerziteta u Beogradu, u okviru seminara Bioinformatika ce biti održano predavanje dr Ane Simonovic sa temom Sekvenciranje genoma i transkriptoma. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Genom predstavlja celokupan genetski materijal jedinke ili vrste. Projekat sekvenciranja jednog genoma, osim samog sekvenciranja, podrazumeva i rekonstrukciju - povezivanje sekvenciranih fragmenata (assembly) i anotaciju – opisivanje razlicitih elemenata genoma i identifikaciju delova koji bi mogli biti geni. Sekvenciranje genoma se može obavljati postupno, strategijom klon-po-klon (BAC-to-BAC Sequencing, Map-Based Sequencing ili Hierarchical Shotgun Sequencing) ili nasumicno (Whole Genome Shotgun Sequencing, WGS). Pored velicine eukariotskih genoma, najveci izazov pri sekvenciranju predstavljajaju repetitivne sekvence, poznate kao junk DNA, kojih može biti daleko više od jedinstvenih sekvenci - samih gena.
Mnogo jednostavniji i jeftiniji, ali i manje informativni su projekti sekvenciranja transkriptoma – skupa informacionih RNK molekula ekstrahovanih iz nekog tkiva. Jedan organizam ima samo jedan genom, koji je isti u svakoj celiji, ali se transkriptomi razlikuju od tkiva do tkiva, jer predstavljaju prepise aktivnih gena u datom tkivu i pod datim uslovima i nikada ne sadrže repetitivne sekvence, ali ni neaktivne gene. Dok su sve jedinstvene genomske sekvence zastupljene ravnomerno u fragmentisanom uzorku genomske DNA, zastupljenost pojedinih transkripata u uzorku ukupne RNK zavisi od nivoa ekspresije gena cije su to kopije. Sekvenciranje transkriptoma se uvek obavlja shotgun pristupom, pri cemu se transkripti fragmentišu, a sekvencirani fragmenti potom sklapaju. Sklapanje transkriptoma se, ako je moguce, obavlja poredjenjem sa sekvencom genoma, a ako genom date vrste nije sekvenciran, rekonstrukcija transkriptoma je moguca i de novo, bez matrice, ali je znatno zahtevnija. Transkriptomika u Srbiji je tek u povoju, a de novo sekvenciranje transkriptoma lekovite biljke kicice (Centaurium erythraeae), koje je u toku, predstavlja pionirski poduhvat iz ove oblasti.
Predavac
Dr Ana Simonovic je rodjena 1969. u Beogradu, gde je završila osnovnu i srednju školu. Diplomirala je na Biološkom fakultetu Univerziteta u Beogradu, na odseku Molekularna biologija i fiziologija 1995., a magistrirala na smeru Fiziologija biljaka 1998. Doktorirala je iz oblasti Molekularna i celijska biologija na Državnom Univerzitetu Severne Dakote, SAD, 2006., kao dobitnik stipendije Presidential Doctoral Fellowship Award. Od 1995. do 2000. i od 2006. do danas je zaposlena na Institututu za biološka istraživanja "Siniša Stankovic" kao naucni saradnik. Angažovana je na osnovnom, tehnološkom i medunarodnom FP7 projektu. Koaturor je više naucnih radova i publikacija iz oblasti biohemije, molekularne biologije i biotehnologije biljaka, kao i autor knjige "Geneticko inženjerstvo i biotehnologija biljaka".
21. decembar 2011, 18h, Matematicki fakultet, sala 718
Novi uvidi u biologiju analizom bioloških mreža
Prof. dr Nataša Pržulj
U sredu, 21. decembra, sa pocetkom u 18 sati, u sali 718 na Matematickom fakultetu Univerziteta u Beogradu, u okviru seminara Bioinformatika ce biti održano predavanje dr Nataše Pržulj sa temom Novi uvidi u biologiju analizom bioloških mreža. Pozivamo sve zainteresovane nastavnike i studente da prisustvuju predavanju.
Apstrakt
Geni proizvode na hiljade razlicitih tipova proteina, koji medusobno reagujuci formiraju kompleksne mreže koje održavaju celijski život. Moguce je da ce podaci o proteinskim interakcijama biti korisni za nova biološka otkrica koliko i genetske sekvence. Zbog velike kolicine interakcijskih podataka, poredenja mreža koje opisuju biološke sisteme patogenih i ne-patogenih vrsta bi mogla imati kljucnu ulogu u razumevanju patoloških mehanizama. Takode, poredenje mreža zdravih i bolesnih celija bi moglo produbiti razumevanje bolesti i voditi identifikaciji celijskih delova koji su kandidati za novu terapeutsku intervenciju.
Postojeci algoritmi za poravnavanje mreža koriste informacije eksterne topologiji mreže, npr. genetske ili proteinske sekvence. Pošto je topologija mreže novi i nezavisan izvor bioloških informacija, važno je razumeti koliko biologije se moze nauciti iz topologije, nezavisno od bilo kog drugog izvora podataka. Zato smo razvili matematicki rigorozne nacine poravnavanja mreža bazirane iskljucivo na topologiji. Naši metodi proizvode do sada najkompletnija poravnavanja mreža koja otkrivaju velike, povezane regione slicnosti izmedu mreža.
Povrh toga, pokazali smo da je topologija mreže drugacija oko gena impliciranih u kanceru i onih koji nisu implicirani u kanceru. Na osnovu toga smo predvideli nove gene potencijalno odgovorne za kancer i naše predikcije su fenotipski validirane. Naše analize pokazuju da graf-teoretske analize bioloških mreža mogu da daju nove uvide u biologiju i filogeniju kao i da pomognu identifikaciji novih proteina za koje se lekovi vezuju, što može doprineti razvoju novih lekova i poboljšanju zdravstvene nege.
Predavac
Prof. dr Nataša Pržulj je vanredni profesor na Racunarskom fakultetu u Beogradu kao i na katedri za racunarstvo Imperial College-a u Londonu. Završila je Matematicku gimnaziju u Beogradu, studirala matematiku i racunarstvo na Beogradskom Univerzitetu, a magistrirala i doktorirala na Univerzitetu u Torontu u Kanadi. Pre dolaska na RAF i Imperial College, radila je kao docent na Univerzitetu u Kaliforniji Irvine. Autor je preko 40 naucnih radova i poglavlja u prestižnim naucnim casopisima i knjigama ukljucujuci Science, Proceedings of the National Academy of Sciences of the USA, Bioinformatics i Bioessays.
7. dec. 2011, 18h, Matematicki fakultet, sala 718
Seminar Bioinformatika - Uvod u molekulsku biologiju (nastavak)
dr Miloš Beljanski, naucni savetnik
U sredu 7. decembra, sa pocetkom u 18 sati, u sali 718 na Matematickom fakultetu u okviru seminara Bioinformatika bice održan drugi deo predavanja dr Miloša Beljanskog sa temom Uvod u molekulsku biologiju.
Veliki broj tema u savremenoj bioinformatici je tesno povezan sa problemima molekulske biologije. Molekulska biologija i grane koje se oslanjaju na nju (na primer genomika i proteomika) proizvode velike kolicine informacija za ciju je obradu potrebno raditi na samim granicama racunarskih nauka. Za uspešno bavljenje ovim oblastima neophodno je poznavanje osnova molekulske biologije. Predavanje dr Miloša Beljanskog ima za cilj da predstavi ovu veoma zanimljivu oblast.
Pozivaju se zainteresovani nastavnici i studenti da prisustvuju predavanju.
23. nov. 2011, 18h, Matematicki fakultet, sala 718
Seminar Bioinformatika - Uvod u molekulsku biologiju
dr Miloš Beljanski, naucni savetnik
U sredu 23. novembra, sa pocetkom u 18 sati, u sali 718 na Matematickom fakultetu u okviru seminara Bioinformatika bice održano predavanje dr Miloša Beljanskog sa temom Uvod u molekulsku biologiju.
Veliki broj tema u savremenoj bioinformatici je tesno povezan sa problemima molekulske biologije. Molekulska biologija i grane koje se oslanjaju na nju (na primer genomika i proteomika) proizvode velike kolicine informacija za ciju je obradu potrebno raditi na samim granicama racunarskih nauka. Za uspešno bavljenje ovim oblastima neophodno je poznavanje osnova molekulske biologije. Predavanje dr Miloša Beljanskog ima za cilj da predstavi ovu veoma zanimljivu oblast.
Pozivaju se zainteresovani nastavnici i studenti da prisustvuju predavanju.
Prezentacija
Odavde možete preuzeti prezentaciju koja je pratila predavanje (PDF).
2. nov. 2011, 18h, Matematicki fakultet, sala 718
Otvaranje seminara Bioinformatika
U sredu 2. novembra, sa pocetkom u 18 sati, u sali 718 na Matematickom fakultetu bice održan prvi sastanak seminara Bioinformatika u školskoj godini 2011/12.
Pozivaju se nastavnici i studenti doktorskih i master studija racunarskih, matematickih, bioloških i drugih srodnih disciplina da iskoriste priliku da se upoznaju sa aktivnostima koje se u ovoj veoma dinamicnoj i atraktivnoj oblasti odvijaju u našoj sredini.
Plan rada
- Predstavljanje Grupe za bioinformatiku (dr Saša Malkov)
- Pregled tekucih aktivnosti (dr Nenad Mitic)
- Razgovor o planovima za naredni period (dr Gordana Pavlovic-Lažetic)
Prezentacija
Odavde možete preuzeti prezentaciju Grupe za bioinformatiku: (PDF).
18 October 2011, 13h, Faculty of Mathematics, University of Belgrade, Room 718
Bioinformatic approaches of studying intrinsic structural disorder of proteins
Prof. Dr. Peter Tompa
On Tuesday 18th October 2011, the Faculty of Mathematics, University of Belgrade, will host the distinguished guest - Professor. Dr. Peter Tompa. Dr. Tompa is director or VIB Department of Structural Biology, Brussels, Belgium, and group leader at Laboratory of Intrinsically Disordered Proteins, Institute of Enzymology, Budapest, Hungary.
Starting at 1pm, in Room 718 (Fourth Floor), Dr. Tompa will give a lecture "Bioinformatic approaches of studying intrinsic structural disorder of proteins". We invite all interested colleagues to attend this lecture.
Abstract
Intrinsically disordered proteins (IDPs/IUPs) exist and function without a well-defined 3D structure, defying the classical structure-function paradigm. Structural disorder is widespread in eukaryotic proteomes and correlates with important functions such as signal transduction and transcription regulation. Whereas currently increasing effort is focused on these proteins, most of our concepts regarding their structure and function stem from a limited range of structural studies. Based on these data, a database of protein disorder (DisProt, www.dispto.org) has been assembled, which is used as a point of reference for developing bioinformatic predictors, i.e. effective computational algorithms to predict disorder from amino acid sequence. In my talk I will briefly overview the field of structural disorder, with a major focus on the special functional modes it enables. I will then discuss the physical principles underlying the most important disorder predictors, and introduce some of the most influential predictors developed. Finally, I will demonstrate through several examples how these tools can be used to address important and diverse biology questions, such as the involvement of structural disorder in cancer, dual coding in the genome and thermal adaptation in bacteria.