PROTEAN

Abstract

The explosion of sequence information has provided a number of candidates for translation into their encoded protein sequences. This primary sequence information provides scant understanding of the function and structure for the novel protein. Evolution has conserved proteins at the functional level; advantageous or neutral functioning proteins continue as disadvantageous proteins are selected against (1). The quaternary and tertiary structure of a protein determines its functionality. This data, unfortunately, can be directly gathered only by NMR or crystallographic studies. In cases in which you have the determined structure of a protein and its primary sequence, you can infer similar structure and function from homologous proteins. When a member of a protein family has been solved, this information can bridge the gap between primary sequence data and tertiary/quaternary data for other members of the family. This holds for identical residue conformations but can diverge rapidly for differing regions. Any predictive information derived from the primary sequence for nonidentity regions may bridge this gap to help build accurate models of structure and inferred functionality. This is especially useful as a first approximation for protein families lacking solved structures. Additionally, primary sequence information can be used to elucidate particular domains of a protein for a more specific application than structural determination, such as locating peptide motifs or antigenic site determination.