Computational statistics in molecular phylogenetics

摘要: Simulation remains a very important approach to testing the robustness and accuracy of phylogenetic inference methods. However, current simulation programs are limited, especially concerning realistic models for simulating insertions deletions (indels). In this thesis I implement new, portable and flexible application, named INDELible, which can be used generate nucleotide, amino acid and codon sequence data by indels (under several models indel length distribution) as well as substitutions rich repertoire substitution models). In particular, introduce study that makes use one INDELible’s many unique features simulate with under codon allow nonsynonymous/synonymous substitution rate ratio vary among sites branches. This is quantify, first time, precise effects alignment errors on false-positive power the widely branch-site test positive selection. Several assessed in context. Through experiment, show do not cause the excessive false positives if correct, but lead to unacceptably high positives. Previous selection studies inferior programs are revisited demonstrate applicability my results in real world situations. Further work uses simulated from INDELible examine tree-shape branch length programs, impact errors on different methods phylogeny reconstruction. analysis performed explore which avoid generating kind most detrimental process of