Biography

Andrei Osterman is an Associate Professor in the Bioinformatics and Systems Biology Program at The Burnham Institute for Medical Research (BIMR) in La Jolla. He received his Ph.D. in Biochemistry at Moscow University in Russia. In 1993 he joined the laboratory of Meg Phillips at the University of Texas in Dallas to perform structure–functional studies of the ornithine decarboxylase enzyme family. Recognizing the new frontiers of metabolic biochemistry and enzymology opened by the genomic revolution, he joined Integrated Genomics, a start-up biotechnology firm in Chicago, in 1999. As a Director and Vice President of Research at this company, he pioneered integration of comparative genomics with biochemical and genetic experiments for gene and pathway discovery. His research team published the first genome-scale study of gene essentiality in E. coli by genetic footprinting. Dr. Osterman is one of the founders of the Fellowship for Interpretation of Genomes (FIG), a nonprofit research organization that launched the Project to Annotate 1,000 Genomes in 2003. FIG provides the open-source integration of all publicly available genomes and tools for their comparative analysis, annotation, and metabolic reconstruction. Dr. Osterman’s laboratory at BIMR focuses on fundamental and applied aspects of the key metabolic subsystems in a variety of species, from bacteria to human. His group applies bioinformatic techniques followed by experimental validation to reconstruct metabolic pathways from genomic data, reveal gaps in their current knowledge, and identify previously uncharacterized (missing) genes. The power of this integrative approach is illustrated by the discovery and characterization of more than twenty enzyme families in the metabolism of cofactors, carbohydrates, and amino acids. Most of the applications pursued by this group are related to pathogenic and environmental bacteria. New research directions include the analysis of regulatory networks and the application of proteomics and metabolomics technology for identification of novel diagnostic and therapeutic targets in cancer.

 

Andrei Osterman , Ph.D.

Associate Professor, Bioinformatics & Systems Biology,
The Burnham Institute for Medical Research, La Jolla, CA
&
Fellowship for Interpretation of Genomes, Bur Ridge, IL

Osterman Lab / Burnham Web
What is Fig? Web

Comparative Genomic Cartography of
Metabolic Networks

Abstract:
We combine comparative genomics with experimental techniques to support the mapping of metabolic networks in hundreds (soon to be thousands) of diverse species with completely sequenced genomes. Dissection of cellular processes to compact subsystems allows us to accurately project functions of individual genes (enzymes, transporters, regulators) and elementary modules (pathways) from model organisms to others. The growing collection of curated subsystems in The SEED genomic integration (http://theseed.uchicago.edu/FIG/index.cgi) also provides us with a framework for interpretation of new types of data, including gene essentiality, proteome and metabolome profiling. Comparative subsystem analysis reveals a stunning picture of conservations and variations in the Central Machinery of Life. Most notably, the observed metabolic diversity appears to emerge, via a combinatorial explosion, from a limited set of operational subsystem variants (stable gene patterns) implemented in different species. Application of genome context analysis techniques, e.g. delineation of conserved operons and regulons in prokaryotes, allows us to efficiently predict functions of novel genes implicated as needed (but missing) for pathway contiguity in certain groups of species. We use focused validation experiments to test the most critical functional predictions. Recent examples will be provided to illustrate applications of this approach for exploration of antiinfective drug targets (NAD synthesis in Francisella tularensis) and for ecophysiological studies of environmental bacteria (carbohydrate utilization in Shewanella oneidensis, osmoprotection in Thermotoga maritima). I will argue that the described approach is scalable into a largely automated annotation-reconstruction-prediction-validation pipeline. Implementing such a pipeline in combination with concerted community effort would rapidly lead to a nearly comprehensive description of metabolic networks in all cultured bacteria. This will set the stage for new breakthroughs in genome-scale metabolic modeling and interpretation of the emerging metagenomic data.