We present the combined application of clustering and composite pattern analysis to Saccharomyces cerevisiae gene expression and sequence data. The aim is to devise a system for the efficient investigation of the upstream regions associated with regulating the coordinated transcriptional response of genes via the location of cis-acting elements. The clustering algorithm presented allows clusters to overlap in light of the fact that many genes are regulated by multiple factors and/or are associated with multiple functions in the organism. The sequence clusters are analysed for shared monad and dyad patterns. These motifs are commonly regarded as putative cis-acting regulatory elements. Computationally, dyad patterns may be difficult and expensive to discover and verify if the sequences are very large. This has highlighted the need to obtain maximally--with regards to the pattern in question--'enriched' sets of sequences to uncover these motifs. We present overlapping clusters as an ideal basis for the discovery of conserved motifs.

• PMID: 15130815
• PubMed
• PubTator

Reference Type

Comparative Study | Evaluation Study | Journal Article | Validation Study

Authors

Magusin A

Primary Lit For

Additional Lit For

Review For