PARSIMONY AND MODEL-BASED PHYLOGENOMIC ANALYSES: A COMPARISON BASED ON EMPIRICAL DATASETS

TORRES, AMBROSIO; CATALANO, SANTIAGO ANDRÉS; GOLOBOFF, PA

Buenos Aires

Encuentro; XXXVI Meeting de la Sociedad Willi Hennig; 2016

The advent of the genomic era in recent years has brought the debate about the suitability of different methods of tree reconstruction into a new stage. Most of the phylogenomic studies published have used model-based methods, possibly driven by the supposed superiority of those approaches over parsimony methods in terms of realism of the models considered. Many empirical and theoretical studies have compared the behavior of parsimony and model-based methods in the past. However, it is not clear how the results of the different approaches differ in the case of phylogenomic analyses where the amount ofdata is impressively higher. For example, the assumption of all characters having correlated branch lengths (i.e. probability of change along a branch) is more and more likely to be violated when a data set includes many different genes, which may lead model-based methods further away from statisticalconsistency than parsimony (which assumes no such correlation). As part of an ongoing project for the comparison of parsimony and model-based methods in phylogenomic studies, we collected 71 phylogenomic matrices (from 2006 to 2016) that included more than 20 000 characters (amino acids or nucleotides). From those we selected the datasets that were analyzed by both Bayesian inference (BI) and Maximum Likelihood (ML). In most of the original analyses MP trees were not calculated, hence we performed the MP searches in TNT. Topological comparisons (SPR-distance and Farris's distortion coefficient) indicated that in 75% of the datasets the topologies found by the different methods -MP, ML and BI- were identical or differ in a single SPR move. On average, ML trees differfrom MP in 0.7 SPR-moves more than they differ from BI trees. In those datasets where the results were more incongruent, the differences were generally associated to nodes that presented low support. In addition the main conclusions drawn in each of the analyses were not affected by those differences. If these preliminary results are confirmed, and considering the computational efficiency of MP methods relative to ML and BI, Parsimony may have an important role in phylogenomics in the future.