TOKYO--Attendees of the Fourth Annual International Conference on Computational Molecular Biology, also known as Recomb 2000, held here April 7-11, were sent home with the challenge to find new routes for bridging experimental biology with computational and technological advances.
Problems such as large-scale genome assembly, analysis of gene networks using expression data and network databases, protein structure analysis, and medical applications for genomics emerged during the four-day event as the field's new frontiers.
Some 320 computer engineers, mathematicians, and computational biologists--nearly 70 percent of them academic--from more than 20 countries attended the meeting. Over 100 scientific posters and 36 papers were presented. The University of Tokyo Human Genome Center organized the event, with sponsorship from the Association for Computing Machinery's Special Interest Group on Algorithms and Computation Theory and several industry, government, and trade organizations.
Papers mainly addressed six current computational biology problems: gene expression and DNA array analysis; links between structure and sequence; construction of phylogenetic trees and development of new algorithms for tracing evolution; technological optimization in DNA assembly from sequences to whole genomes; comparative genomics; and structure prediction by combining data from sequence and structure.
The meeting featured eight invited speakers.
Minoru Kanehisa of Kyoto University discussed graphical and chemical approaches to analysis of interacting molecules using the metabolic pathway database KEGG. He observed that having genes and their protein products is only a step toward the greater goal of describing relationships among proteins and viewing them as interacting units.
Hans Lehrach, of the Max Planck Institute in Berlin, explained a new approach, the Oligonucleotide FingerPrint method, for finding novel or homologous genes by large-scale gene expression analysis using model organisms.
Suante Paabo of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, presented surprising findings on diversity and nucleotide variations within the human genome, and shared his comparison of DNA sequence variation among 70 humans and 30 chimpanzees.
Eric Davidson from California Institute of Technology gave a talk on sea urchin development, demonstrating the need for computational analysis for the modeling of cis-regulatory control systems at various stages.
Takashi Gojobori of Japan's National Institute of Genetics in Mishima presented an analysis of complete genome sequences from three kingdoms to find differences in stability among genome structures.
Walter Gilbert of Harvard University discussed the role, origin, and future of introns, emphasizing the hidden information that links introns' evolution and protein structure and function. He expressed the importance of theoretical analysis to guide experimental works and database analysis to find novel genes in expanding data sources.
Leroy Hood of the Institute for Systems Biology in Seattle, Wash., emphasized biology as an information science. Classical biology will shortly be replaced by a global systems biology, he contended, which aims to understand life through the modeling of genes, proteins, pathways, interaction of components, and interaction of networks.
Not all human genomes are the same, emphasized Yusuke Nakamura, a researcher at the University of Tokyo who described Japan's effort to genotype its population using state-of-the-art technologies. Nakamura addressed personalized medicine--prevention, diagnosis, and therapy for individuals according to their genotypes--and showed, as an example, a test case in which colon cancer patients could be effectively treated according to their individual gene expression profiles.
Recomb 2001 will be held in Montreal, April 21-24.
--Michal Linial and Toru Yao