ENZYME-SPECIFIC PROFILES for metabolic pathway prediction
PRIAM: PRofils pour l'Identification Automatique du Métabolisme
Clotilde CLAUDEL-RENARD, Claude CHEVALET, Thomas FARAUT, Daniel KAHN
Clotilde Claudel-Renard, Claude Chevalet, Thomas Faraut and Daniel Kahn / Enzyme-specific profiles for genome annotation: PRIAM Nucleic Acids Research, 2003, Vol. 31, No. 22 6633-6639
The prediction of metabolic pathways of many species is becoming possible with the advent of complete sequence genomes, on the basis of the identification of enzyme coding genes.
PRIAM is a method for automated enzyme detection in a fully sequenced genome, based on all sequences available in the ENZYME database.
PRIAM relies on sets of position-specific score matrices (PSSMs) automatically tailored for each ENZYME entry. The whole Swiss-Prot database has been used to parameterise and to assess the method.
As an example, PRIAM was applied to predict metabolic pathways from the complete genome of the nitrogen fixing bacterium Sinorhizobium meliloti and on the plant pathogen Ralstonia solanacearum. The results of this automated annotation method were compared with the original genome annotation and visualised on KEGG graphs in order to facilitate the interpretation of predicted metabolic pathways and to highlight potentially missing enzymes.
Release of July 2006 (Enzyme Release 39 28-Jun-2006) -> download)
Information on profiles of PRIAM
( EC for example: 126.96.36.199 )
Search for a protein
RPSBLAST with PRIAM (jul 2006)
( Choose for database: profil_ENZYME_39 )
Complete proteome analysis package
PRIAM was designed initially to assist annotators in the intepretation of the metabolism of an organism on the basis of its complete genome sequence. PRIAM was meant to indicate whether any particular enzyme is likely to be encoded in the genome, without emphasis on individual genes. This implied sets of rules determining which modules should be present in order to infer the presence of each enzyme. Although it is possible to trace results back to individual genes, this approach does not necessarily identify all relevant subunits, particularly in the case of enzyme complexes.
It has now appeared that several researchers also use PRIAM profiles for annotation of individual genes, a slightly different usage. This raises different issues: for example an ancillary protein might be found, even in the absence of the corresponding catalytic unit.
In order to facilitate the latter usage, we now provide two versions of PRIAM:
- The genome-oriented release uses rules as published in the original paper. These rules operate on a subset of selected modules.
- The gene-oriented release comprises representative profiles matching all the polypeptides referenced in the ENZYME database. They can be searched using RPS-BLAST. Beware however that these profiles do not necessarily correspond to catalytic modules!
For information, comments and / or suggestions on PRIAM,
please email us at PRIAM Team