Skip to main content
. 2021 Feb 10;38(6):2446–2467. doi: 10.1093/molbev/msab038

Fig. 2.

Fig. 2.

Depth of taxonomic sampling is more influential than matrix completeness in supermatrix analyses of pseudoscorpion placement. (a) Internal phylogeny of Pseudoscorpiones showing major taxonomic groups. Notations on nodes indicate taxon subsets obtained by sequential pruning of branches. (b) Sensitivity plot of 42 phylogenomic matrices assembled by varying gene occupancy (y axis) and taxonomic sampling (x axis), with partitioned model-fitting. Colors of squares correspond to the sister group of Pseudoscorpiones obtained in each maximum likelihood analysis. (c) Analysis of largest (1002 loci) and densest (248 loci) matrices under variations of the PMSF model. (d) Selected tree topologies showing the dynamics of pseudoscorpion instability as a function of taxonomic sampling in partitioned-model analyses.