Software

Fast-Find: A novel computational approach to analyzing combinatorial motifs

Micah Hamady¹ , Erin Peden² , Rob Knight³ and Ravinder Singh²

¹Department of Computer Science, University of Colorado, Boulder, CO 80309, USA

²Department of Molecular, Cellular and Developmental Biology, University of Colorado, Boulder, CO 80309, USA

³Department of Chemistry and Biochemistry, University of Colorado at Boulder, Boulder, CO 80309, USA

author email corresponding author email

BMC Bioinformatics 2006, 7:1doi:10.1186/1471-2105-7-1

Published:	4 January 2006

Abstract

Background

Many vital biological processes, including transcription and splicing, require a combination of short, degenerate sequence patterns, or motifs, adjacent to defined sequence features. Although these motifs occur frequently by chance, they only have biological meaning within a specific context. Identifying transcripts that contain meaningful combinations of patterns is thus an important problem, which existing tools address poorly.

Results

Here we present a new approach, Fast-FIND (Fast-Fully Indexed Nucleotide Database), that uses a relational database to support rapid indexed searches for arbitrary combinations of patterns defined either by sequence or composition. Fast-FIND is easy to implement, takes less than a second to search the entire Drosophila genome sequence for arbitrary patterns adjacent to sites of alternative polyadenylation, and is sufficiently fast to allow sensitivity analysis on the patterns. We have applied this approach to identify transcripts that contain combinations of sequence motifs for RNA-binding proteins that may regulate alternative polyadenylation.

Conclusion

Fast-FIND provides an efficient way to identify transcripts that are potentially regulated via alternative polyadenylation. We have used it to generate hypotheses about interactions between specific polyadenylation factors, which we will test experimentally.