Computational challenges in structural and functional genomics

The goal of computational biology in the early twenty-first century is to link the various genome sequencing projects to a high-throughput effort in complete structural and functional annotation of whole genomes or biological pathways. It is, in fact, a logical extension of the genome effort to systematically elaborate DNA (deoxyribonucleic acid) sequences into full three-dimensional structures through to functional analysis of cellular networks. The first level of the biological hierarchy is comparative analysis of the rapidly emerging genomic data at the sequence level. However, knowing only the sequence of DNA does not always tell us about the structure or function of the genes, nor does it tell us about the combined action of their protein products, which is the essence of higher order biological function. Complete annotation will include the determination of structure and function of proteins, and a move from analysis of these individual macromolecules to their complex interactions that make up the processes of cellular decisions. This paper represents an effort by a research community to define the hard computational biology problems of the future, to define what mixture of basic research directions and practical algorithmic approaches will be required to achieve our goals, and to outline the directions that will likely be taken in the postgenomic era.