Biological Sequence Analysis

The central problems of biological sequence comparison are the definition of measures of sequence similarity, the development of algorithms for optimizing these measures, and the description of statistics for evaluating sequence similarity significance. Our previous research has provided the algorithmic and statistical underpinnings of widely used methods for sequence analysis, including the BLAST and PSI-BLAST database search programs. Continuing work focuses on the improvement of the similarity measures and statistics used by these and related programs, with the aim of improving the detection of subtle biological relationships.
Stephen Altschul, Ph.D.

Comparative Analysis of Protein Structure

The growing data base of known three-dimensional structures contains a wealth of information on molecular evolution. Research focuses on methods to extract and utilize this information efficiently, and to apply it to problems in molecular recognition and structure prediction. Further research interests include the following areas: (1) software tools to facilitate the biologist's access to three-dimensional structure data and comparative analysis results, (2) algorithms for detecting and classifying structural similarity, and (3) algorithms for sequence structure threading and fold recognition.
Stephen Bryant, Ph.D.

Reconstruction of Organismal Biology Using Protein Sequence and Structure Analysis

Elucidating the biochemical and biological functions of protein domains is central to our understanding of life processes. Our research uses in silico methods, based on evolutionary principles, for the discovery, classification and biochemical predictions of protein domains. We also study the evolution of functionally relevant biomolecule complexes and networks comprised of ensembles of proteins, and reconstruct such functional assemblages using various techniques of comparative genomics.
Aravind Iyer, Ph.D.

Evolutionary Genomics

The rapid growth of the database of sequenced genomes from all walks of life creates unprecedented opportunities for reconstructing ancestral life forms and understanding the forces that affect genome evolution at different levels. Our research addresses a variety of themes in evolutionary genomics, from attempts to reconstruct the earliest events in life’s evolution to elucidating the selective processes acting at different types of sites in mammalian genomes.
Eugene Koonin, Ph.D.

Computational Molecular Biology of Chromosomal Proteins, Nuclear Organization, and Gene Regulation

The DNA of all eukaryotic cells is organized in chromatin so that all nuclear processes are capable of functioning accurately. Although there are multiple orders of compaction of the DNA, the packaging is not random, and it is also capable of maintaining such functional mechanisms as gene transcription regulation, replication, and mRNA processing. Our research centers around computational aspects and modeling of the protein-protein and protein-DNA interactions, as well as the DNA secondary structure, with specific reference to nucleosomes, chromatin higher structure, and transcription.
David Landsman, Ph.D.

Computational Molecular Biology/Biological Sequence Analysis

Research interests includes (1) the design and integration of data bases for genetic and physical maps of complex genomes, DNA and protein sequences, and protein structures and motifs; (2) the development of tools for the access and analysis of biological data (generally sequence data); and (3) the study of molecular evolution, the development of appropriate measures of similarity for biological sequences, and the application of existing tools to the elucidation of macromolecular function.
David Lipman, M.D.

Systems Biology

The functions of a biological system largely depends on the mutual interactions among its constituent components such as genes or their products in a metabolic or signaling pathway. New experimental techniques provide a vast amount of data on protein-protein interactions, genetic interactions, and other interactions at the molecular level. Analysis and modeling of this data provides a unique perspective on cell function. Our interest in this field includes the analysis of protein interaction networks to identify functional modules, the prediction of protein and domain interactions, the analysis and modeling of signaling pathways, and the analysis transcriptional networks.
Teresa Przytycka, Ph.D.

Mathematics and Statistics in Bioinformatics

The group maintains a continuing interest in mathematics and statistics relevant to bioinformatics, in particular the local maximum (segmental exceedance) statistic and related topics in renewal theory. Other areas interest include accurate p-values for statistical tests in bioinformatics, including sequence and structure comparisons, and the foundations of evaluating database retrieval algorithms. Biological interests include the prediction of transcription factor binding elements and the evaluation of repeats.
John Spouge, M.D. Ph.D.

Techniques for Optimizing Textual Information Retrieval

Current research focuses on Medline text retrieval as an information retrieval problem, with emphasis placed on optimizing methods in this context. Both Bayesian and vector methods are under investigation. New algorithms are being designed that place heavy emphasis on pattern recognition at the phrase level, and that use machine learning strategies to augment their ability to recognize textual objects. Individuals with a background and interest in machine learning and free text information retrieval problems are encouraged to apply.
John Wilbur, M.D. Ph.D.

Computational Approaches to Problems of Malaria

Our computational research on the biology of malaria and related parasites includes (1) genome-wide quantitative genetic analysis of drug resistance and other phenotypes; (2) recombination and genomic diversity in parasite populations; (3) antigen evolution; (4) novel physical modeling methods for protein domains with low complexity, repeated, or disordered sequences related to pathogenicity and host immune responses; and (5) evolution and function of extreme compositional bias in parasite genomes and proteins. Some projects are in close collaboration with consortia of experimental laboratories; others involve fundamental exploratory concept development motivated by parasite biology and disease control strategies. Opportunities are available for candidates with computer programming ability and backgrounds in quantitative/mathematical aspects of genetics and evolution, or in physics, physical chemistry, or macromolecular biophysics.
John Wootton, Ph.D.

Biological Statistical Physics and Bioinformatics

Although governed by the principles of physics, a biological system usually consists of more degrees of freedom than that of a simple physics system where complete analysis is feasible. Because of the diversity of elementary components involved, straightforward use of statistical physics is also not suitable. We combine first principle physics, such as electromagnetism and quantum mechanics, statistical physics, and bioinformatics to study biological systems. We use both the top-down and bottom-up approaches. For example, we use bioinformatics analysis to analyze the problem of the patterns such as transcription factor binding sites, followed by statistical physics and electrostatics analysis. To understand the biomolecular interaction in a complex environment, we reformulate electrostatics to provide a computational scheme that uses controllable approximations, i.e. the accuracy of the computation is tunable depending on computational resources. This new formulation can easily be applied to study protein- protein interactions and protein-RNA structures. Another area of interest in our group is proteomics. We have been developing a computational method, RAId, to sequence peptides and identify proteins using tandem mass spectrometry data.
Yi-Kuo Yu, Ph.D.

Revised: October 1, 2009