|
|
| |
| |
Dr. Peer BORK
Senior Group Leader (Bioinformatics) & Joint Coordinator,
Structural and Compuational Biology unit,
EMBL Heidelberg, Germany
bork@embl.de
|
About the Speaker:
Peer Bork, PhD, is senior group leader and joint coordinator of the Structural and Computational Biology unit at EMBL, a European research organization with headquarters in Heidelberg. He also holds an appointment at the Max-Delbrueck-Center for Molecular Medicine in Berlin. Dr Bork received his PhD in Biochemistry (1990) and his habilitation in Theoretical Biophysics (1995). He works in various areas of computational biology and systems analysis with focus on function prediction, comparative analysis and data integration. He published more than 350 research articles in international, peer-reviewed journals, among them 40 in Nature, Science or Cell. According to ISI Dr. Bork is currently the most cited European researcher in Molecular Biology and Genetics. He is on the editorial board of a number of journals including Science and PloS Biology, and functions as senior editor of the journal Molecular Systems Biology. Dr Bork co-founded four biotech companies, two of them went public.
More than 20 of his former associates hold now professorships or other group leader positions in prominent places all over the world.
|
Abstract:
Predicting Biological Function at Different Spatial Scales
|
Biological systems range from small biomolecules to entire ecosystems. We use conceptually similar comparative approaches to uncover functional aspects of diverse biological systems at different spatial scale. I will illustrate our approach with examples from metazoan genome annotation, from small-molecule-protein network analysis (e.g. how to use side effect to predict drug targets) and from environmental genomics (e.g. how to discover metabolic differences at millimeter scale).
|
Dr. Francesca D. CICCARELLI
Group Leader,
European Institute of Oncology (IEO)
Francesca.Ciccarelli@ifom-ieo-campus.it
|
About the Speaker:
Francesca Ciccarelli, PhD, is group leader at the European Institute of Oncology (IEO) a leading research Institute in cancer biology, located in Milan, Italy. Dr Ciccarelli graduated at the University of Bologna in 1998 and worked at the Istituto Mario Negri (Italy) until 2000. After that, she moved to the EMBL in Heidelberg, where she joined the group of Dr. Bork (2001-2005). In 2003 she earned the PhD at the University of Heidelberg and in July 2005 started her own group at the IEO.
The main scientific interest of Dr. Ciccarelli is studying the interplay between the forces that govern genome evolution and those leading to genomic instability during cancer development. In particular, she aims at defining systems-level properties of genes involved in cancer. She is also interested in rebuilding the evolution of cancer genes through the comparison of genome sequences.
|
Abstract:
Low Duplicability and Network Fragility of Cancer Genes
|
Large heterogeneity has been recently reported in the number and types of genes that can undergo driver genetic modifications during cancer development. Finding the biological basis of this heterogeneity is instrumental for a better understanding of the entire tumorigenic process. During my talk, I will explain a combination of genomic and network-based approaches that we applied to identify systems-level biological properties of a group of 350 well-known cancer genes. We also extended our analysis to around 250 candidate cancer genes identified through large-scale unbiased mutational screenings. We found that, regardless of their molecular function, cancer genes retain significantly less duplicates than other human genes. In addition, they code for protein hubs occurring within highly interconnected modules of the human protein-protein interaction network.
Comparable genomic and network properties recur also within candidate cancer genes, while they differ significantly from those of singleton human genes not involved in cancer. Our study shows that cancer genes are particularly fragile components of the human gene repertoire, sensitive to dosage modification. It also contributes to explain the heterogeneity of cancer genetics and support the interpretation of tumor as a systems disease.
|
Prof. Martijn A. HUYNEN
Principal Investigator, Nijmegen Centre for Molecular Life Sciences
Group Leader, Comparative Genetics, Centre for Molecular & Biomolecular Informatics
Radboud University Nijmegen Medical Centre
M.Huynen@cmbi.kun.nl
|
About the Speaker:
Martijn A. Huynen (1964) obtained his MSc degree in 1993 in Bioinformatics at Utrecht University with P. Hogeweg. From 1993 to 1996 he worked as a postdoc at the Los Alamos National Laboratory and the Sante Fe Institute, New Mexico, USA on the evolutionary dynamics of RNA and the prediction of its secondary structure. From 1996 to 2001 he worked at the European Molecular Biology Lab in Heidelberg, Germany in the group of Peer Bork. During his time at the EMBL he started his work on comparative genome analysis to compare pathways and predict protein function, being one of the first people to publish about the potential of comparing genomes to predict protein function and pathways. In 2001 he came to the Radboud University Nijmegen Medical Center as a group leader, becoming full professor in 2002.
He leads a group of about 10 people that are active in various areas of comparative genomics, with an emphasis on method development to extract reliable and specific predictions about protein function and pathways from genomics data and on the evolution and workings biomolecular systems like the mitochondria and the immune system. He has numerous collaborations with experimental groups for the testing of predicted protein functions and pathways and for the analysis of genomics data.
|
Abstract:
Tracing the Evolution of the Mitochondrial Proteome and Predicting its Function
|
The wealth of sequenced genomes in combination with rapidly increasing knowledge about biomolecular systems us to 1) trace the evolution of these systems and 2) predict their functions. The presentation will specifically focus on mitochondria. 1: Mitochondria are eukaryotic organelles that originated from the endosymbiosis of an alpha-proteobacterium. By comparing alpha-proteobacteria with current day mitochondria, we delineate the evolution of the mictochondrial proteome. Overall, there has been a large turnover of the mitochondrial proteome in evolution. Early on, proteins involved in cell envelope synthesis have disappeared, whereas proteins involved in replication, transcription, cell division, transport, and regulation have been replaced by eukaryotic proteins. More than half of what remains from the mitochondrial ancestor in modern mitochondria is involved in translation and in energy conversion.
The results indicate that the eukaryotic host has hijacked the proto-mitochondrion, taking control of its protein synthesis and metabolism. 2: Current day mitochondria come in all shapes and sizes, from minimal genome-less mitosomes whose only function appears to be the assembly of iron-sulfur clusters, to full fledged mitochondria with complete respiratory chains. By mapping this variation on a phylogenetic tree and examining the co-evolution of proteins we predict the function of new mitochondrial proteins. Several of these predictions have been verified experimentally.
|
Dr. Jan KORBEL
Group Leader, Gene Expression unit
EMBL Heidelberg, Germany
jankorbel75@gmail.com
|
About the Speaker:
My PhD research in Peer Bork's group at EMBL Heidelberg (2001-2005) focused on computational biology, in particular gene function prediction and protein interaction network inference using genomic context analysis. In 2005, funded by an EMBO Longterm fellowship as well as a Marie Curie Outgoing International fellowship, I moved to the USA for postdoctoral research. In particular, I began working in Mark Gerstein's and Michael Snyder's groups at Yale University in functional genomics, i.e. using a hybrid approach that involves both computational analyses and applying experimental high-throughput technologies. The biological focus of my most recent research has been in studying human genome variation - in particular genome structural variants - using novel high-resolution genomics technologies and computational data mining.
Recently, I decided to return to Heidelberg, Germany, to take a position as an EMBL Group Leader starting in October 2008. I am presently already spending much at EMBL, where I am working as a visitor in Peer Bork's group, finishing up postdoctoral research projects, and planning my new research group.
|
Abstract:
Global Mapping of Genome Structural Variation in Humans Using Functional Genomics Technologies
|
JO Korbel, AE Urban, J Affourtit, B Godwin, F Grubert, J Simons, PM Kim, D Palejev, N Carriero, L Du, B Taillon, Z Chen, A Tanzer, E Saunders, J Chi, F Yang, N Carter, M Hurles, SM Weissman, T Harkins, MB Gerstein, M Egholm and M Snyder
Genome Structural Variants (SV), involving large, kilo- to mega-base sized deletions, duplications, insertions, inversions, and complex combinations of rearrangements, are widespread in the genomes of healthy individuals and presumably responsible for a considerable amount of phenotypic variation. Until recently, cost-effective methods for large-scale analysis of SVs have identified variants in the order of 50kb or greater, but have failed to detect smaller SVs or to precisely identify boundaries of SVs (i.e. SV breakpoints. I will present an approach that we recently developed, high resolution and massive Paired-End Mapping (PEM), which combines rescue and capture of the paired-ends of 3 kb fragments, next-generation DNA sequencing, and a computational method that involves mapping of massive sets of DNA reads onto the human reference genome, to identify SVs 3 kb or larger. The resolution of PEM is considerably higher than that of previous approaches.
Using over 30 million paired-ends we recently mapped SVs in an African and putatively European individual and identified shared and divergent SVs relative to the reference genome (Korbel et al. Science 2007). Overall, we fine-mapped more than 1000 SVs and documented that the number of SVs among humans is much larger than initially hypothesized; many potentially affect gene function. To systematically analyze the physical boundaries of SVs, we inferred breakpoint junction sequences for more than 200 by DNA sequencing and computational analysis. Systematic analysis of breakpoint sequences revealed specific classes of SVs and mechanisms of SV formation. SVs were mostly formed by non-homologous end-joining, retrotransposition, and non-allelic homologous recombination involving high-frequency repetitive elements or low copy repeats. In some cases polymorphic deletion events evidently caused gene fusions, resulting in novel genes that appear to be present in a portion of the human population.
Overall our study revealed an unprecedented number of SVs in humans and provided insights into mechanisms by which SVs have arisen in the genome. Together with an international consortium, we are presently using PEM and other approaches to chart a global high-resolution map of SVs in different human populations.
|
Asst. Prof. Shamil SUNYAEV
Assistant Professor of Medicine and Health Sciences and Technology,
Division of Genetics, Department of Medicine,
Brigham & Women's Hospital, Harvard Medical School
ssunyaev@rics.bwh.harvard.edu
|
About the Speaker:
Shamil Sunyaev received his PhD from Moscow Insitute of Physics and Technology (MIPT) and completed postdoctoral training at European Molecular Biology Laboratory (EMBL). He is now an Assistant Professor at Genetics Division, Brigham & Women's Hospital, Harvard Medical School. He is also a member of Harvard-M.I.T. Health Sciences and Technology Division. His interests include computational analysis of human genetic variation, comparative genomics and computational proteomics.
|
Abstract:
Predicting Pathogenicity of Missense Mutations
|
Rapid development of sequencing technology puts to the forefront the need to interpret DNA sequence information. Sequencing of phenotyped clinical populations is widely anticipated to replace genotyping in studies aiming at finding genes underlying human complex diseases. Clinical genetic diagnostics by sequencing is increasingly important in guiding therapeutic intervention and providing counseling to family members of patients with monogenic and oligogenic diseases, although interpretation of results of diagnostic sequencing is sometimes problematic because sequencing of patient cohorts discovers many variants of unknown significance (VUS). These developments create an unprecedented demand for development of computational methods for interpretation of nucleotide and amino acid changes, and for predicting the effect of mutations and polymorphism on molecular function, fitness and phenotype.
We developed and extensively tested a new method PolyPhen2 for predicting the functional effect of human missense mutations. The method combines comparative sequence analysis, analysis of protein 3D structure and database annotations. We created a new highly accurate automated multiple sequence alignment pipeline and designed several novel features predictive of the effect of mutations. We employed a number of machine learning techniques to select the best set of features and to generate prediction rules. These developments resulted in a greatly increased sensitivity and specificity of predictions as evident from tests on several datasets. We also analyzed applicability of the method in the setting of a clinical genetic diagnostics lab.
|
Dr. Chandra VERMA
Head, Biomolecular Modelling and Design division
Principal Investigator, Atomic Stimulations and Design in Biology group
Bioinformatics Institute, A*STAR
chandra@bii.a-star.edu.sg
|
About the Speaker:
Chandra Verma received his doctorate in York UK in 1990. He subsequently joined the structural biology laboratory in York, applying techniques of molecular simulations to understand links between protein structures, dynamics and functions. In 2003 he joined the Bioinformatics institute in Singapore as a group leader where he currently leads the Biomolecular Modelling & Design division. The group is involved in establishing dynamical models of protein functions and in peptide and drug design.
|
Abstract:
Computational Discovery of New Pathways and Their Modulations
|
Detailed bioinformatic analyses have been used to discover novel biological pathways involving lipid post-translational modifications that modulate processes in biology and the discovery of a novel pathway will be discussed. At a more detailed level of representation, that of molecules, data will be presented that will show how simple and careful analyses using physics-based structural modelling provide insights that have enabled the discovery of new interactions in the p53 pathway. Both these examples were subsequently confirmed experimentally and highlight the role computational biology is increasingly playing in providing hypotheses that have lead to new discoveries in biology.
|
Prof. Christian von MERING
Group Leader, Bioinformatics Group
Institute of Molecular Biology, University of Zurich
mering@molbio.uzh.ch
|
About the Speaker:
Prof. von Mering studied biochemistry at the Free University of Berlin. He earned his Ph.D. at the University of Zurich, working in Developmental Biology with Prof. Dr. Konrad Basler. He then joined the group of Dr. Bork at the EMBL in Heidelberg as a Postdoctoral Fellow. His work focuses on the following areas: protein network analysis, data integration, and evolution. Together with Peer Bork, Dr. von Mering was one of the first to realize the shortcomings of raw high-throughput interaction measurements, but also their potential to reveal functional modularity in protein networks. He has focused onimproving such networks by filtering experimental datasets against each other, and by developing algorithms for scoring and visualization. Prof. von Mering also has experience with various other types of high-throughput data, notably in the area of genome sequencing and comparative genomics, where he focuses on integrating data from several model organisms.
Since 2002, he has published 33 peer-reviewed articles in the fields of bioinformatics and network biology (including one paper in Nature, and four papers in Science), and has achieved an ISI h-Index of 17. He is a member of the Research Priority Program 'Systems Biology' at the University of Zurich, and has an additional group leader affiliation at the Swiss Institute of Bioinformatics in Lausanne. He is a member of the editorial board of PLoS Computational Biology."
|
Abstract:
Comparative Model Organism Proteomics - What It Can Tell Us About The Evolution of Protein Abundance Levels
|
Similar to the genomics revolution, systematic shotgun proteomics is now beginning to yield organism-wide datasets that vastly increase our knowledge about the cellular machinery. These data reflect the expression status of entire proteomes, under a given experimental setting. Since this type of data is currently being produced for a number of model organisms, we can begin to study the evolution of (core) proteomes. Which parts of the proteome are consistently expressed, and at what levels? How does protein expression correlate with genomic organization, and what protein modifications might be conserved? I will discuss two of the largest proteomics datasets to date, one from D.melanogaster, and one from C.elegans.
These have specific strengths and weaknesses, and I will present a strategy for extracting quantitative information about protein expression status. By mapping this to orthology information and transcript abundance data, we can begin to characterize the evolutionary forces shaping the transcriptome.
|
Prof. Limsoon WONG
Professor & Vice Dean (Research), School of Computing, National University of Singapore (NUS)
Professor, School of Medicine, NUS
Leader, Bioinformatics Programme, NUS Office of Life Sciences
Coordinator, Computational Biology Lab, NUS School of Computing
wongls@comp.nus.edu.sg
|
About the Speaker:
Limsoon Wong is a professor in the School of Computing and the School of Medicine at the National University of Singapore. He is currently working mostly on knowledge discovery technologies and is especially interested in their application to biomedicine. He serves on the editorial boards of Journal of Bioinformatics and Computational Biology (ICP), Bioinformatics (OUP), and Drug Discovery Today (Elsevier). He received his BSc(Eng) in 1988 from Imperial College London and his PhD in 1994 from University of Pennsylvania.
|
Abstract:
Drug Pathway Decipherer
|
Existing computational tools focus on pathway enrichment analysis by identifying informative genes which are differently expressed between two response groups, but little information on the interplay between selected genes is provided. The identifications are too general and hardly sufficient to generate specific hypotheses for our research purpose. I plan to describe a drug pathway identification system ---the Drug Pathway Decipherer (DPD) ---developed by my student Difeng Dong. The DPD generates hypotheses of specific genetic pathways based on the knowledge of canonical biological pathways, which promises the identifications to be properly interpreted in a biological context. I plan to illustrate the DPD using a case study on nasopharyngeal carcinoma.
|
|
|
