QMaker: Fast and accurate method to estimate empirical models of protein evolution

• DOI: 10.1101/2020.02.20.958819
• Publication Date: 2020-02-23T19:15:12Z
• Authors: Bui Quang Minh; Cuong Cao Dang; Le Sy Vinh; Robert Lanfear
• Keywords: Substitution (logic); Tree (set theory); R package

Abstract Amino acid substitution models play a crucial role in phylogenetic analyses. Maximum likelihood (ML) methods have been proposed to estimate amino models, however, they are typically complicated and slow. In this paper, we propose QMaker, new ML method general time-reversible Q matrix from large protein dataset consisting of multiple sequence alignments. QMaker combines an efficient tree search algorithm, model selection for handling the heterogeneity among alignments, consideration rate mixture sites. We provide as user-friendly function IQ-TREE software package ( http://www.iqtree.org ) supporting use CPU cores so that biologists can easily their own used empirical current Pfam database well five clade-specific mammals, birds, insects, yeasts, plants. Our results show considerably improve fit between data some cases influence inference topologies.