 |
 |
 |
 |
 |
 |
|
|
||||||
|
 |
The School of |
|
|
|
Invited SpeakersLiming Cai Associate Professor, University of Georgia "RNA structural homology search with stochastic grammar models" Runsheng Chen Professor, Bioinformatics Laboratory, Institute of Biophysics, Chinese Academy of Sciences. "Conservation analysis of small RNA genes in Escherichia coli" Bailin Hao "BGF --- The Beijing Gene Finder" Fuchu He Professor, Beijing Institute of Radiation Medicine, member of the Chinese Academy of Science. "Bioinformatics Analysis for the Proteome of Human Fetal Liver Aged 22 Weeks of Gestation" Jenn-Kang Hwang Professor, Institute of Bioinformatics, National Chiao Tung University, HsinChu "The relationship between structural entropy and protein thermostability" Luhua Lai Professor, College of Chemistry and Center for Theoretical Biology, Peking University "Functional Protein Design based on Protein-Protein Interaction" Hao Li Assistant Professor, Department of Biochemistry & Biophysics, University of California, San Francisco. "Genomic Reconstruction of the Transcription Networks of A Cell" Songgang Li Professor, Center for Bioinformatics, Peking University "Discovering Novel Targets from the Human Genome" Yixue Li "Protein-protein Interaction Analysis" Zhi-Rong Sun Professor, Tsinghua University "In Silico Identification of the Key Components and Steps in IFN-gamma Induced JAK-STAT Signaling Pathway" Wang Wei and Jing Li "Amino acid grouping and protein sequence complexity" Zhiping Weng Associate Professor, Biomedical Engineering Department, Boston University "Computational Approaches to Molecular Interaction" Dong Xu Associate Professor, Computer Science Department, University of Missouri, Columbia. "Predicting Rho-independent terminators from bacterial genomic sequences" Ying Xu Professor, Biochemistry and Molecular Biology Department and the Institute of Bioinformatics, University of Georgia (UGA) "Computational methods for protein mass spectral data interpretation" Hongyu Zhao Associate Professor, Public Health and Genetics, Yale University "Integrated Statistical Modeling of Gene Expression Data" Chun-Ting Zhang Professor, Tianjin University, Member of the Chinese Academy of Sciences "Isochore structures in eukaryotic genomes" Michael Q. Zhang Professor, Cold Spring Harbor Laboratory "Combinatorial Regulation of Transcriptional Gene Networks" Yaoqi Zhou Assistant Professor, State University of New York at Buffalo "Making physical interactions out of statistics of protein" Liming Cai Title: RNA structural homology search with stochastic grammar models Abstract: RNA secondary structure remains to be the best hope for an exploitable characteristics of ncRNAs. Indeed, an increasing number of structural homology search tools, mainly based on stochastic context-free grammars (SCFGs), have recently been developed for ncRNA gene identification. In contrast to the Tinoco's thermodynamics model, SCFGs can easily include statistical biases that often occur in RNA sequences, making it more suitable to profile specific RNA structures for structural homology search. Profile SCFGs are usually large in size, however, making model constructions difficult and search algorithms inefficient. In this presentation, a succinct SCFG model is introduced for profiling homologous RNA secondary structure. It is demonstrated with supporting test data that not only does the new model makes profile construction effective and homology search efficient, but also it allows a simple extension of the SCFG to include pseudoknotted structures which were too complex for SCFGs to model. Liming Cai received his B.S. and M.S. in computer science from Tsinghua University in 1984 and 1986 respectively, and Ph.D. in computer science from Texas A&M University in 1994. Having taught at East Carolina University and Ohio University, he joined University of Georgia in 2001 where he is currently an associate professor in computer science. His research interests include bioinformatics and computational biology with focuses on probabilistic and linguistic models for the structural aspects of biological sequences and on the efficiency of algorithms developed based upon such models. Runsheng Chen Title: Conservation analysis of small RNA genes in Escherichia coli Abstract:Small RNA (sRNA) genes have drawn much attention in later years because of their abundance and diversity. Studies of sRNA have focused on Escherichia coli K-12 because of its important role as a model organism and 44 out of 55 experimentally confirmed sRNA genes have been precisely located in the genome. The object of this study is to analyze quantitatively the conservation of these sRNA genes and compare it with the conservation of protein-encoding genes, function-unknown regions and tRNA genes. The results show that within an evolutionary distance of 0.26, both sRNA genes and protein-encoding genes display a similar tendency in their degrees of conservation at the nucleotide level. In addition, the conservation of sRNA genes is much stronger than function-unknown regions, but much weaker than tRNA genes. Based on the conservation of studied sRNA genes, we also give clues to estimate the total number of sRNA genes in E.coli. Bailin Hao Title: BGF --- The Beijing Gene Finder Abstract: A team was formed in August 2001 at the Hangzhou Branch of BGI to write a gene-finder from scratch. It was primarily aimed to predict genes in the rice genome. BGF is based on SHMM (Semi-Hidden Markov Model) and Dynamic Programming, but many minor or major problems and adjustments have to be made in the process of testing. We have constructed two test datasets for rice using the 28000 odd cDNA published by Kikuchi et al in July 2003: a single-gene set of 500 sequences and a multi-gene set of 46 sequences with 199 genes in total. We tried our best to diminish potential overlap of the test datasets with the training datasets used by us and by other gene-finders for rice. We ran six gene-predicting progarms on the test datasets: BGF, FgeneSH, GeneMark, GenScan (trained for maize), GlimmerR and RiceHMM. For the time being BGF and FgeneSH are the best ones for predicting genes in the rice genome. However, our goal was not to rate the performance of the programs but rather to explore the main factors affecting the gene predictions in all these programs based on probabilistic models. The website of BGF is: http://rise.genomics.org.cn:8080/ and it is open to public for pasting or uploading their genomic sequences for gene prediction. At present it works for the rice, fruitfly, and silkworm genomes. Fuchu He Title: Bioinformatics Analysis for the Proteome of Human Fetal Liver Aged 22 Weeks of Gestation Abstract: In the human physiological activities, liver is not only the largest gland in metabolism of carbohydrates, fats, proteins, vitamins and hormones, but also an organ of secreting bile, detoxifcation, devourment, and protection function. The human fetal liver aged 22 weeks of gestation (HFL22w) is the major source of stem/progenitor cells of embryonic hematopoiesis and immune system. It also has a lot of genes relating to cellular recognition, migration, and homing. The study of the human fatal liver will help us understanding liver development and its physiological functions. For this reason, the protein expression profile of fetal liver aged 22 weeks of gestation is analysed by bioinformatics approaches in five aspects as below. (1)Function classification of proteins (2)Protein interaction network (3)Getting the full length cDNAs and MS identification of unknown genes (4)Analysis of protein subcellular localization (5)Integration and comparison of its transcriptome and proteome Fuchu He, PhD, is academician of the Chinese Academy of Science. Dr. He received his B.S. degree in Genetics from Fudan University, Shanghai in 1982. Then he earned his M.S degree in Biochemistry and Ph.D. in Cell Biology from Beijing Institute of Radiation Medicine. He is currently the Professor of Beijing Institute of Radiation Medicine and the Director of China National Center of Biomedical Analysis. Based on his research work, Dr. He found several periodic phenomena:¡°Development-related evolution¡± of cytokines; ¡°Co-evolution¡± of cytokines and their receptors; ¡°Modulated evolution¡± of mRNA coding regions and their non-coding regions, etc. During the research in human fetal liver, his team discovered, cloned, termed hepatopoietin (HPO) from human fetal liver, and demonstrated and characterized its features. At the same time, he discovered two important related pathways via its two receptors. To search more important factors from fetal liver, his lab established the world¡¯s largest scale and the most systematic gene-expression profiles for liver and fetal liver. In addition, his research team discovered, cloned and identified more than 100 human novel genes, most of which have been confirmed and characterized, listed by international authoritative data banks and published or/and patented. In our country, Dr. He took the lead in sparkpluging and developing the proteomics research actively. He is the chief scientist of the National ¡°973¡± Project and the Key Technologies Research and Development programme. Recently, he, as the chief scientist, initiated a series of the Chinese Human Proteome Program, and then international HUPO Human Liver Proteome Project (HLPP). He has been nominated as the Chair of HLPP Committee. Jenn-Kang Hwang Title: The relationship between structural entropy and protein thermostability Abstract: We developed a technique to compute the structural entropy directly from protein sequences. We explored the possibility of using structural entropy to identify residues involved in thermal stabilization of various protein families. Our results showed that the positions of the largest structural entropy differences between the wild type and the mutant usually coincide with the residues relevant to thermostability. We also observed a good linear relationship between the average structural entropy and the melting temperatures for each protein family. The linear relationship between structural entropy and protein thermostability should be useful in the study of protein thermal stabilization. Prof. Jenn-Kang Hwang Chairman, Department of Biological Science & Technology, Director, Institute of Bioinformatics, National Chiao Tung University, HsinChu Luhua Lai Title: Functional Protein Design based on Protein-Protein Interaction Abstract: Protein-protein interaction plays major roles in biological regulatory networks and can be used as a starting point for functional protein design or structural based drug design. Compared to conventional one-protein and one-ligand approach, protein networks and pathways have been paid more attention. We have studied the characteristics of the protein-protein interface and developed a method to graphically analyze protein interface. In order to quantitatively calculate binding free energies of protein complexes, we have built a statistical potential and successfully applied it in prediction protein binding free energies. An abnormal type of hydrogen-bond, which involves hydrogen atom attached to a carbon atom, was found to play significant role to stabilize protein complex. Studies towards protein-protein interaction provide basis for function protein design. In order to understand the structure and function relationship and to design functional proteins with potential applications, we have established a strategy to graft functional residues, which are discontinuous in sequence, onto possible protein skeletons. EPO-EPOR system was used as an example to design new proteins with epo activity. Luhua Lai, full professor at College of Chemistry and Center for Theoretical Biology, Peking University. Professor Lai¡¯s group works on protein-protein, protein-ligand interactions, protein structure and function prediction, protein design, structural based drug design, and signal transduction pathways. B.Sc., 1984, Peking University. M.Sc., 1987, Peking University. Ph.D., 1989, Peking University. 1998-1999, Berkeley Scholar, University of California at Berkeley. Email: lhlai@pku.edu.cn; http://ctb.pku.edu.cn Telephone: 010-62757486,62757520 Fax: 010-62751725 Hao Li Title: Genomic Reconstruction of the Transcription Networks of A Cell Abstract: Organisms devote a significant fraction of their genomes to encoding regulatory information which specifies when and where different genes should be turned on or off. Such information is essential for understanding development, tissue specificity, and cellular response to the environment and has great importance for understanding the molecular basis of disease. The regulatory programs encoded in DNA are executed by transcription factors which respond to different conditions and regulate gene expression combinatorially, leading to complex regulatory networks. We have developed theoretical models and computational algorithms to decipher the regulatory programs and to reconstruct the transcription networks in model organisms. Our approach consists of the following steps: 1) systematically identifying the DNA recognition sites and target genes of transcription factors in the genome; 2) inferring environmental and genetic perturbations under which each transcription factor is activated/deactivated; 3) analyzing combinatorial regulation by multiple transcription factors. Our approach integrates information from the genome sequences and genome-wide gene expression data, and combines bioinformatics analysis with mechanistic models of gene regulation. Assistant Professor, Department of Biochemistry & Biophysics, University of California, San Francisco. Songgang Li Title: Discovering Novel Targets from the Human Genome Abstract: Discovering novel targets from the human genome remains an important problem. We present an overview of an integrated platform that we have developed, streamlining construction of databases of known target families, integration of pathway databases, gene prediction, and prediction of novel targets. First, we evaluated four popular gene-finding programs, GenScan, GrailII, GeneID, and HMMGENE, and implemented a new method that combines the four programs to increase sensitivity and specificity. Using the new method, we predicted over 100,000 exon segments which had resulted in 6000 novel genes. We have developed several databases of known target families such as GPCR, nuclear receptor, and secreted proteins. We also integrated the information on gene-pathway association from nine pathway databases into one searchable pathway database. Using the above data, we had predicted secreted proteins and genes related to apoptosis in the human. Some of the novel secreted proteins have already been validated since our prediction. We mapped genes predicted to be in a target family to the human genome, and found interesting evidence of gene clusters. We are in the process of validating some of our predictions in a wet lab. Songgang Li is a Professor at the Center for Bioinformatics in Peking University. He had previously studied system ecology and risk assessment of genetic modified organism. In recent years he had focused on bioinformatics methods for the sequencing and analysis of whole genomes, and played key roles in the draft sequencing and assembly of the rice indica genome. His current research interests are in the study of genes in pathways at the genome level, and the incorporation of microarray data in the recognition of pathways. He aims to integrate multiple sources of data and tools, using both in silico and experimental approaches, in the prediction of novel targets. Yixue Li Title: Protein-protein Interaction Analysis Abstract: Protein interaction underlie most cellular functions. Protein-protein interactions result in the formation of transient or stable multi-subunit complexes so called protein machine. An understanding of these complexes is required for the functional annotation of proteins and is a step towards the elucidation of molecular pathways such as signaling cascades and regulatory networks. Now a lot of bioinformatics methods are useful for protein-protein interactions analysis, for example, in sillico two hybrid, mirror tree, gene fusion, etc. Now we use some method, such as SVM, improved phylogenetic profiling, for the analysis of protein-protein interactions. Good results were gotten, and it can give also an elucidation of molecular pathways. Further these methods will be used for building protein-protein interaction map for some model system. Zhi-Rong Sun Title: In Silico Identification of the Key Components and Steps in IFN-gamma Induced JAK-STAT Signaling Pathway Abstract: We address the issue of global identification of the key components and steps in signal transduction networks at a systems level. This is investigated using a mathematical model of IFN-gamma induced JAK-STAT signaling pathway by applying robustness analysis and multi-parametric sensitivity analysis based on a large number of Monte-Carlo simulations and statistical evaluation of simulation results. Specifically, we performed global sensitivity analysis on the mathematical model by systematically varying the initial concentrations of the proteins and the kinetic parameters values of this signaling pathway. We demonstrate that the JAK-STAT signaling pathway is robust with respect to (w.r.t.) its 'response time' but not w.r.t. 'response magnitude'. Our study further indicates that protein levels of SOCS1, cytoplasmic STAT1, and nuclear phosphatases are important for the dynamical systems behaviors of this pathway and that the critical processes of this signal transduction are nuclear import and export of STAT1 rather than the upstream processes. Zhirong Sun obtained his BS from Tsinghua University in 1970. Now He is a full professor of Biology Department, Tsinghua University. In the early years of Zhi-Rong Sun's research, he was engaged in the study of Control and Informatics, After 80s he shifted to the area of theoretical molecular biology. In 90s, extend to Bioinformatics. He has been a head of Bioinformatics Lab. Dong Xu Title: Predicting Rho-independent terminators from bacterial genomic sequences Abstract: Rho-independent terminator (RIT) is a stretch of DNA sequence, whose RNA transcript forms a stem-loop structure to terminate transcription without the Rho protein involved. Studying RIT is important for defining operons and understanding gene transcription in bacteria. To predict RITs from bacterial genomic sequence, we developed a novel algorithm called Rnall (RNA Local secondary structure prediction by Local symmetry mapping). Rnall first extracts the possible local secondary structure candidates from RNA transcripts based on dynamic programming and then uses energy minimization to optimize local secondary structure. We can obtain a linear time complexity O(N) with a space requirement O(N) when scanning RITs in a genome, where N is the length of genome sequence. We applied Rnall to screen RITs in E. coli and all of 130 known RITs from RegulonDB were successfully retrieved. We carried out a systematic study of RITs for Synechococcus sp. based on Rnall in conjunction with comparative genomic studies using 7 other related cyanobacteria whose complete genomic sequences are available. Multiple sequence alignment and phylogenetic footprinting of the RITs in different genomes reveal interesting evolutionary patterns in conservation and divergence. Our study is useful for understanding transcription termination and improving the accuracy of operon prediction and metabolic pathway prediction in cyanobacteria. Dr. Dong Xu is a James C. Dowell Associate Professor and Director of Digital Biology Laboratory in the Computer Science Department, University of Missouri, Columbia. He is also a part-time faculty of U.S. Department of Energy Higher Education Research Experience at the Oak Ridge National Laboratory for Distinguished Visiting Scientists and Faculty. He got his Ph.D. from the University of Illinois, Urbana-Champaign in 1995 and did two-year postdoctorial work at National Cancer Institute. He was a Staff Scientist at Oak Ridge National Laboratory until August 2003 before joining University of Missouri. Over the past thirteen years, he has been involved in many areas of computational biology and bioinformatics with more than 70 scientific papers. He is the recipient of the year 2001 R&D 100 award, a prestigious international award sponsored by Research & Development magazine that honors the 100 most significant new technical products of the previous year, for developing "Protein Structure Prediction and Evaluation Computer Toolkit (PROSPECT)". He also received 2003 Award of Excellence in Technology Transfer from The Federal Laboratory Consortium for the developing gene expression analysis package EXCAVATOR. Ying Xu Title: Computational methods for protein mass spectral data interpretation Abstract: In this talk, I will present a new computational framework for interpretation of mass spectral data for protein identification. The framework is based on a novel method for separation of different ion types in tandem mass spectral data. The ion-type identification problem is formulated as a graph partition problem and solved rigorously using efficient algorithms. We then extend this method to solve the de novo sequencing problem and the problem of identification of post-translational modifications, key challenging problems in computational proteomics. Ying Xu is a chair professor under the title "Regents-Georgia Research Alliance Eminent Scholar" in bioinformatics and computational biology, in the Biochemistry and Molecular Biology Department and the Institute of Bioinformatics, University of Georgia (UGA). Before joining UGA in Sept 2003, he was a senior staff scientist and group leader at the Oak Ridge National Laboratory (ORNL), where he still holds a joint position. He also holds guest or research professor positions at the University of Tennessee at Knoxville, Jilin Uiniversity, and Zhejiang University of China. He received his Ph.D. degree in theoretical computer science from the University of Colorado at Boulder in 1991. His Ph.D. thesis work was on development of efficient algorithms for matroid intersection problems (supervised by Hal Gabow). Between 1991 and 1993, he was a visiting assistant professor at the Colorado School of Mines. He started his bioinformatics career in 1993 when he joined Ed Uberbacher's group at ORNL to work on the GRAIL project. Since then, he has been doing research and development work in various areas of bioinformatics and computational biology, ranging from computational genomics, proteomics to computational systems biology. His current research interests include (a) protein structure prediction and modeling, (b) large-scale biological data mining, (c) computational inference of biological pathways, and (d) cancer bioinformatics. He is interested in both development of computational techniques for biological problems and investigation of biological problems using in silico approaches. He has published over 100 research papers in the open literature, including two books in bioinformatics (MIT Press, 2002) and genomics (John Wiley and Sons, March 2004). He has also given over 80 invited/contributed talks at various conferences, workshops, research organizations and universities. He has (co)developed a number of bioinformatics software, including GRAIL II & GRAIL EXP, PROSPECT and EXCAVATOR. He enjoys teaching and interacting with students. He co-taught a graduate-level bioinformatics course in 2002, and will co-teach an undergraduate-level bioinformatics course in the Spring of 2004. Over the years, he has been actively involved in various bioinformatics conferences and journals. In 2003-2004, he is the Program Committee (co)Chair of the IEEE Computer Society Bioinformatics Conference (CSB'04), and a Subject-area Chair in the Program Committee of the joint ISMB/ECCB conference in 2004. He currently serves on the editorial boards of three international journals. He has also served on various review panels for major funding agencies, scuh as NSF, NIH and DOE. Wang Wei and Jing Li Title: Amino acid grouping and protein sequence complexity Abstract: Finding the relationship between amino acids and simplifying the folding alphabets are important for understanding protein folding and for designing new proteins. Here in this work, we put forward a new clustering method and apply to two different Protein databases with different sequences similarities. We use relative entropy analysis, the homology detection and the principal component analysis to find the relationship of the 20 letters and to obtain the new substitution matrice. Our results show that our method can well apply to clustering amino acids, especially when the sequences similarities are low. Furthermore, we make sequence alignment with the simplified alphabets. By comparing to the protein structure alignment, we find that the sequence alignment can well match the structure alignment when 10 letters are used. Zhiping Weng Title: Computational Approaches to Molecular Interaction Abstract: In this lecture I describe two projects in our lab: 1. prediction of cis-elements for gene regulation. In essence this is due to protein-DNA interaction; 2. prediction of the 3-dimensional (3D) of protein-protein complex structures from the 3D structures of the input proteins. Dr. Weng graduated from the University of Science and Technology of China in 1992 with B.S. in Electrical Engineering. In 1993, she entered the graduate program in Biomedical Engineering at Boston University, and received her Ph.D. in 1997. The focus of her thesis research was in computational biology, especially in protein-protein docking. In January 1997 Dr. Weng was appointed Instructor of Biomedical Engineering at Boston University. In that capacity she taught and conducted research, and had primary responsibility for the development of the Bioinformatics program and the core curriculum in Bioinformatics. In January 1999 the Biomedical Engineering Department at Boston University decided to grow in the area of Bioinformatics. After a national search, the department appointed Dr. Weng a tenure-track assistant professor. In September 2003, Dr. Weng was promoted to Associate Professor with tenure. Dr. Weng's research is focused on developing computational methods to obtain a predictive understanding of protein-protein interaction and transcriptional regulation. Her lab currently has 3 postdocs, 12 graduate students, funded by the National Institutes of Health, the National Science Foundation, and the Whitaker Foundation. She has published 47 archival journal articles. Dr Weng has developed a graduate course titled "Protein and DNA Sequence Analysis", which is one of the core courses in the Bioinformatics program. She has taught the course every fall since 1997. In addition to students from the Biomedical Engineering Department and the Bioinformatics Program, the course attracts many students from other universities and researchers in biotech companies. Chun-Ting Zhang Title: Isochore structures in eukaryotic genomes Abstract: The isochore structure is a basic organization of many eukaryotic genomes and it also has important biological implications, e.g., isochores have been correlated with chromosome bands, gene density, recombination rate, replication timing. Although isochores were found long time ago based on experimental evidence, with the availability of the genome sequences of many eukaryotic genomes, such as the human genome, the identification of isochores become even more difficult, largely due to the lack of suitable techniques. We proposed a windowless method for the GC content computation, which is termed the cumulative GC profile. Based on this method, we have identified 56 isochores in the human genome, 28 isochores in the mouse genome and 15 isochores in the Arabidopsis thaliana genome. We further show that the rat genome also has clear isochore structures. High resolution isochore boundaries have also been determined. We proposed a new index to measure the homogeneity of the isochore GC content. The biological significance of the identified isochores is discussed. Chun-Ting Zhang graduated from the department of physics, Fudan university in 1961 and finished his postgraduate study in the same university in 1965. Now he is a full professor of Tianjin university. In the early years of Chun-Ting Zhang's research, he was engaged in the study of theoretical physics. Later, he shifted to the area of theoretical molecular biology. Chun-Ting Zhang was elected to be a Member of the Chinese Academy of Sciences in 1995, and a Fellow of the Third World Academy of Sciences (TWAS) in 2001. Hongyu Zhao Title: Integrated Statistical Modeling of Gene Expression Data Abstract: Recent advances in large-scale RNA expression measurements, DNA-protein interactions, and the availability of genome sequences from many organisms have opened the opportunity for massively parallel biological data acquisition and integrated understanding of the genetic networks underlying complex biological phenotypes. Many existing statistical procedures have been proposed to analyze a single data type, e.g. clustering algorithms for microarray data and motif finding methods for sequence data. However, different data sources offer different perspectives on the same underlying system, and they can be combined to increase our chance of uncovering underlying biological mechanisms. In this talk, we will describe our attempts to develop a statistical framework to integrate diverse genomics and proteomics information to dissect transcriptional regulatory networks. The developed methods will be illustrated through their applications in the reconstruction of trascription networks during yeast cell cycle. This is joint work with Ning Sun, Baolin Wu, and Liang Chen. Dr. Hongyu Zhao is the Ira V. Hiscock Associate Professor of Public Health and Genetics at Yale University. He received his B.S. in Probability and Statistics from Peking University in 1990 and Ph.D. in Statistics from the University of California at Berkeley in 1995. Between 1995 and 1996, he was an Adjunct Assistant Professor at UCLA, and has held visiting positions at the National Cancer Institute and Harvard University School of Public Health. He was awarded the Evelyn Fix Memorial Medal and Citation by UC Berkeley and a Basil O'Connor Starter Scholar Award by the March of Dimes Foundation. His research interests are the applications of statistical methods in molecular biology and genetics, including (1) identification of genetic variants underlying complex diseases; (2) biological network modeling and analysis; and (3) disease biomarker identification through proteomics. He has published over 80 research papers on many research areas, including statistics, human genetics, bioinformatics, proteomics. Dr. Zhao currently serves on the editorial boards of several journals, and is a member of the Biostatistical Methods and Study Design Study Section at the National Institutes of Health. He has served on review panels for various federal funding agencies and private foundations. More information about his research activities can be found at http://bioinformatics.med.yale.edu. Michael Q. Zhang Title: Combinatorial Regulation of Transcriptional Gene Networks. Abstract: Transcriptional regulation in eukaryotes is known to occur through the coordinated action of multiple transcription factors (TFs). Studying combinatorial regulation by traditional biochemical and genetic experiments has been difficult. Recently, a large number of genome-wide experimental data has become available; it has become a great challenge for computational biologists to develop new algorithms that can integrate such a diverse and noisy data for extracting underlying biological functions. In this talk, I will discuss several novel computational approaches that intend to integrate three functional genomics data sources: genome sequence, ChIP and microarray expression and focus on identification of cooperative transcription factors and their binding sites. Michael Q. Zhang, professor of Watson School of Biological Sciences, Cold Spring Harbor Laboratory and at the adjunct Faculty of SUNY at Stony Brook in the Genetics Program and Biomedical Engineering Program. He is also a Guest Professor at Tsinghua University. He serves as a regular member of the NIH grant review panel and as an acting associate editor of Bioinformatics and an editor of Nucleic Acid Research. His research interests include computational genomics and bioinformatics. Yaoqi Zhou Title: Making physical interactions out of statistics of protein structures. Abstract: Understanding of how proteins fold, bind, and/or function requires an accurate energy function to describe water-mediated interaction between amino acid residues. A simple, effective method to obtain an approximate energy function is to extract it from known structures of proteins (called knowledge-based statistical potential). However, different compositions of amino acid residues at the core, the surface, and the binding interface of proteins prohibited the establishment of a common physically-meaningful statistical potential for folding and binding studies despite the fact that the physical basis of the interaction is the same. Here, we show how a simple physical principle can make a knowledge-based potential physically more accurate and lead to a significant leap in the accuracy for selecting native structures of protein and protein-protein complexes from decoys and predicting loop conformation, stability changes upon mutations, and binding free energy of protein-protein/peptide complexes. Moreover, the new energy function provides a quantitative understanding for the burial of hydrophobic residues in protein interior. 1990, Ph.D. in Chemical Physics, SUNY at Stony Brook, Advisors: Harold Friedman and George Stell. 1990-1993 Applied Phys. & Chem. Laboratory, Scientist/Director. 1994-1995 Postdoc at Chemical Engineering, North Carolina State University (with Carol Hall). 1995-2000 Postdoc at Chemistry, Harvard University (with Martin Karplus). 2000-2004, Assistant Professor, State University of New York at Buffalo. |
