|
Abstract:
|
An essential problem in comparative genomics is to identify functional elements through conservation analysis. Phylo-HMM was recently introduced to detect conserved elements based on multiple genome alignments, but usually it was used without strict statistical performance evaluation. We investigated the performance of phylo-HMM by simulation. In order to mimic the true promoter alignment, the parameters estimated from the true promoter alignment were used to simulate the data. It showed the performance of this approach depends on many factors beside the correctness of the model per se, such as the number of genomes used, the phylogenetic tree, signal abundance, expected signal length and the sequence evolution model. The results gave hints about where and how to use this method, and what is the performance we can expect.
|
