A new phylogenetic protocol: Dealing with model misspecification and confirmation bias in molecular phylogenetics

Lars S Jermiin; Renee A Catullo; Barbara R Holland

Then follows the task of selecting sites that will be used to infer the phylogenetic tree. The rationale behind doing so is to maximize the signal-to-noise ratio in the MSA. By omitting poorly-aligned and highly-variable sections of the MSA, which are thought to create noise due to the difficulty of establishing true homology for each site (defined as similarity due to historical relationships by descent (49)), it is hoped that the resulting sub-MSA will retain a strong historical signal (defined as the order and timing of divergence events (50)) that will allow users to obtain an accurate phylogeny. The choice of sites to retain is made by visual inspection of the MSA or by using purpose-built software (51-61). The automated ways of filtering MSAs have been questioned (62).