PhySpeTree: an automated pipeline for reconstructing phylogenetic species trees

Abstract

BackgroundPhylogenetic species trees are widely used in inferring evolutionary relationships. Existing software and algorithms mainly focus on phylogenetic inference. However, less attention has been paid to intermediate steps, such as processing extremely large sequences and preparing configure files to connect multiple software. When the species number is large, the intermediate steps become a bottleneck that may seriously affect the efficiency of tree building.ResultsHere, we present an easy-to-use pipeline named PhySpeTree to facilitate the reconstruction of species trees across bacterial, archaeal, and eukaryotic organisms. Users need only to input the abbreviations of species names; PhySpeTree prepares complex configure files for different software, then automatically downloads genomic data, cleans sequences, and builds trees. PhySpeTree allows users to perform critical steps such as sequence alignment and tree construction by adjusting advanced options. PhySpeTree provides two parallel pipelines based on concatenated highly conserved proteins and small subunit ribosomal RNA sequences, respectively. Accessory modules, such as those for inserting new species, generating visualization configurations, and combining trees, are distributed along with PhySpeTree.ConclusionsTogether with accessory modules, PhySpeTree significantly simplifies tree reconstruction. PhySpeTree is implemented in Python running on modern operating systems (Linux, macOS, and Windows). The source code is freely available with detailed documentation (https://github.com/yangfangs/physpetools).

Benefits of alignment quality‐control processing steps and an Angiosperms353 phylogenomics pipeline applied to the Celastrales

“…In phylogenomic analyses, alignment is often relegated to an automated pipeline (e.g., Wu and Eisen, 2008; Dunn et al., 2013; Fang et al., 2019) and manual examination is considered impractical (Wu and Eisen, 2008) and unsustainable (Edwards et al., 2016). But there is a middle path wherein one identifies loci with highly incongruent gene trees relative to those inferred from other loci and then targets them for manual examination (Simmons et al., 2016; Springer and Gatesy, 2018).…”

Wiring Between Close Nodes in Molecular Networks Evolves More Quickly Than Between Distant Nodes

“…We obtained concatenated alignments of 27 highly conserved protein sequences for the six strains using PhySpeTree ( Fang et al 2019 ). We simultaneously estimated the phylogeny and divergence times in BEAST2 ( Bouckaert et al 2014 ) using the following approach: We employed one partition for each protein, with linked trees/clocks.…”

Wiring between close nodes in biological networks evolves more quickly than between distant nodes

“…Table 2. We built a maximum likelihood phylogenetic tree of the six strains using PhySpeTree (Fang et al 2019) to obtain concatenated alignments of 35 highly conserved protein sequences (HCP). We use this concatenated alignment of HCP sequences to construct a bayesian phylogenetic tree in BEAST2 (Bouckaert et al 2014) using a birth-death tree with a strict molecular clock, and calibration nodes from the literature to estimate divergence times between species (Battistuzzi et al 2004).…”

“…We obtained concatenated alignments of 27 highly conserved protein sequences for the six strains using PhySpeTree (Fang et al 2019). We simultaneously estimated the phylogeny and divegence times in BEAST2 (Bouckaert et al, 2014) using the following approach: We employed one partition for each protein, with linked trees/clocks.…”