Target selection for structural genomics

Abstract

Bioinformatics analyses provide invaluable guidance in finding the shortest path to the ultimate goal of structural genomics – full understanding of structure–function relationships in proteins and exhaustive exploration of the sequence‐structure space. Advanced computational tools allow for predicting currently intractable target proteins and those that already have a known fold, grouping sequences into families, and propagating available structural information to a large number of protein sequences by means of homology modeling. Rational target selection is also a major cost‐saving factor because it prevents the international structural genomics consortia from duplicating effort and helps to identify the most economical set of target proteins. The main strategies in target selection include the systematic determination of protein structures from completely sequenced genomes of selected model organisms, elucidation of all fold types existing in nature, and exhaustive covering of the entire sequence space with structural information.