The PIR S uper F amily ( PIRSF ) classification system

Abstract

Protein family classification provides effective means for large‐scale genome annotation and biological knowledge discovery based on the information embedded within families of homologous sequences and their structures. The Protein Information Resource (PIR) developed the PIRSF (SuperFamily) system, a network classification system based on evolutionary relationship of full‐length proteins , to facilitate the propagation and standardization of protein annotation. Sequence analysis and protein classification based on full‐length proteins, including the preservation of domain architecture, can lead to educated predictions for both generic biochemical and specific biological functions. The multiple levels of sequence diversity, from superfamilies to subfamilies, reflect different degrees of functional granularity and, thereby, allow more accurate propagation of annotation and the development of standard protein nomenclature and ontology. The PIRSF database is integrated with other family, function, and structural classification schemes, and is accessible at http://pir.georgetown.edu/pirsf/ for report retrieval and sequence classification.

Large‐scale, classification‐driven, rule‐based functional annotation of proteins

“…Pending confirmation of activity, "thiamine diphosphate-dependent enzyme" would be a safer annotation. As these examples demonstrate, the use of a hierarchical database helps prevent over-or underannotation (see Barker et al, 2005, …”