Finding consensus patterns in very scarce biosequence samples from their minimal multiple generalizations

Abstract

In this paper we examine the issues involved in finding consensus patterns from biosequence data of very small sample sizes, by searching for so-called minimal multiple generalization (mmg), that is, a set of syntactically minimal patterns that accounts for all the samples. The data we use are the sigma regulons with more conserved consensus patterns for the bacteria B. subtilis. By comparing between the mmgs found over different search spaces, we found that it is possible to derive patterns close to the known consensus patterns by simply making some reasonable requirements on the kinds of patterns to obtain. We also propose some simple measures to evaluate the patterns in an mmg. © Springer-Verlag Berlin Heidelberg 2006.