Diego Darriba

ModelTest-NG is a faster and more feature-rich tool for selecting DNA and protein evolutionary models, with improved accuracy and new functionalities.

RAxML-NG improves maximum likelihood phylogenetic inference with better accuracy, speed, and features like terrace detection and transfer bootstrap support.

The paper introduces EPA-O_SCPLOWNGC_SCPLOW, a faster and more scalable phylogenetic placement algorithm for metagenomic data, outperforming existing methods by up to 30 times in sequential mode.

The paper assesses code quality in evolutionary biology software and highlights issues with floating point arithmetic, proposing best practices and advocating for better funding and policy support.

The paper tests if nucleotide substitution models affect tree topologies using ML inference on 39 datasets, finding that model selection can lead to substantial topological differences in about 5% of cases.

This study evaluates partitioning strategies in phylogenomic datasets, finding that statistical methods and a priori model assignments improve accuracy, while unpartitioned data reduce tree quality.