Superfamily classification in PIR-international protein sequence database

Abstract

Publisher Summary This chapter discusses the superfamily classification in the Protein Information Resource (PIR)-International protein sequence database and the development of the model of the protein superfamily concept that encompasses the most common usages and integrates both homology at the domain level and homology at the level of complete proteins. This model preserves the ability to fully partition the Protein Sequence Database, permits the organization of the database in a structured way, and introduces a more precise and unambiguous definition for the term “protein superfamily.” This classification provides a systematic scheme for the verification of the information in the database and for inferring additional information by homology in a controlled way. Information generated from large-scale sequencing projects is incomplete and not well understood. The major task of computational biology is to assign biological meaning to these data. Homology is the major operating principle employed in these analyses. The superfamily classification provides a useful architecture for the self-consistent and objective examination of sequence data by homology.