Abstracts

Genetic Bases of Biodiversity

Pattern databases have become vital tools for identifying distant relationships in sequences, and hence for predicting protein function and structure. Software tools and a new method have been designed to construct patterns of protein families. By using such method, databank of protein family patterns, PROF_PAT, has been produced. Motifs of patterns, which had the minimum level of probability to be found in random sequences, are selected. Flexible fast search program accompanies the bank. The researcher can specify a similarity matrix (PAM, BLOSUM and other). Variable levels of similarity can be set (permitting search strategies ranging from exact matches to increasing levels of "fuzziness"). Thousands of amino acid sequences can be treated in a few minutes.

To exploit the potential of the PROF_PAT bank of patterns to the full, it is crucial to develop PROF_PAT on the basis of the most recent versions of the protein database SWISS-PROT + TrEMBL. An automatized procedure is developed. Each new version of PROF_PAT is created during a month after new version of SWISS-PROT or TrEMBL is appeared. The current version of this bank is based on 320,000 complete sequences of SWISS-PROT (rl.39) and TrEMBL (rl.16), and contains patterns for more than 27,000 groups of related proteins in a format similar to that of the PROSITE. The number of motifs is 431062, their length is 4241824, comparing to more than 57,000,000 positions in proteins of PROF_PAT families.

The bank is placed on the molecular genetics server of Institute of Cytology and Genetics of the Russian Academy of Sciences (http://wwwmgs.bionet.nsc.ru/mgs/programs/prof_pat/, Novosibirsk, Russia), its local version is is available via FTP ( ftp://ftp.bionet.nsc.ru/pub/biology/vector/prof_pat/, ICG, Novosibirsk, Russia), and (ftp://ftp.ebi.ac.uk/pub/databases/prof_pat/, European Bioinformatics Institute, Hinxton, Great Britain).

Note. Abstracts are published in author's edition

|Home Page| |English Part|

Go to Home