A New Algorithm for Identifying Cis-Regulatory Modules Based on Hidden Markov Model

Figure 2

Correlation of coassociated motifs. In gene regulation, a TF usually works synergistically with other TFs by interactions to regulate a highly specific expression pattern. To facilitate such interactions, their binding sites are located adjacent to each other and form modules, also called composite elements (CEs) [3]. The same CEs perform similar functions in different genes, and they should be conserved in sequences and have a preferred arrangement. To capture coassociated motifs constituting CEs, the model defines the correlation probabilities. Based on CE’ conservation assumptions, such motif pairs may cooccur repeatedly in regulatory sequences of genes, which are marked in rectangles in the example. The region within each pair of brackets represents a CRM. Each polygon represents a motif site within a CRM.