SIMPLE SEQUENCE REPEATS IN THE HELICOBACTER-PYLORI GENOME

We describe an integrated system for the analysis of DNA sequence moti fs within complete bacterial genome sequences. This system is based ar ound ACeDB, a genome database with an integrated graphical user interf ace; we identify and display motifs in the context of genetic, sequenc e and bibliographic data, Tomb et al, (1997) previously reported the i dentification of contingency genes in Helicobacter pylori through thei r association with homopolymeric tracts and dinucleotide repeats. With this as a starting point, we validated the system by a search for thi s type of repeat and used the contextual information to assess the lik elihood that they mediate phase variation in the associated open readi ng frames (ORFs), We found all of the repeats previously described, an d identified 27 putative phase-variable genes (including 17 previously described), These could be divided into three groups: lipopolysacchar ide (LPS) biosynthesis, cell-surface-associated proteins and DNA restr iction/modification systems, Five of the putative genes did not have o bvious homologues in any of the public domain sequence databases, The reading frame of some ORFs was disrupted by the presence of the repeat s, including the alpha(1-2) fucosyltransferase gene, necessary for the synthesis of the Lewis Y epitope. An additional benefit of this appro ach is that the results of each search can be analysed further and com pared with those from other genomes, This revealed that H. pylori has an unusually high frequency of homopurine:homopyrimidine repeats sugge sting mechanistic biases that favour their presence and instability.