Messenger RNA Polyadenylation Site Recognition in Green Alga Chlamydomonas Reinhardtii

摘要

Recognition of polyadenylation [poly(A)] sites for messenger RNA is important in genome annotation and gene expression regulation analysis. In the paper, poly(A) sites of Chlamydomonas reinhardtii were identified using an updated version of poly(A) site recognition software PASSJVAR based on generalized hidden Markov model. First, we analyzed the characteristics of the poly(A) sites and their surrounding sequence patterns, and used an entropy-based feature selection method to select important poly(A) signal patterns in conservative signal states. Then we improved the existing poly(A) sites recognition software PASS that was initially designed only for Arabidopsis to make it suitable for different species. Next, Chlamydomonas sequences were grouped according to their signal patterns and used to train the model parameters through mathematical statistics methods. Finally, poly(A) sites were identified using PASS_VAR. The efficacy of our model is showed up to 93% confidence with strong signals.