Directory list:

dataProvenance.xlsx: contains the accession numbers and sources for the proteomic and transcriptomic data used in this study

rawData.zip: contains the pre-processed input data

• OrthoFinder_input contains the raw genomes/proteomes/transcriptomes listed in dataProvenance.xlsx
• orthoGroups_raw.zip contains the raw OrthoFinder output, excluding the WorkingDirectory and the empty Single_Copy_Orthologue_Sequences
• orthoGroups_paraFilter contains the 183 paralogue filtered orthogroups, see main text for methodology

scripts.zip: contains any additional script needed to run/modify analyses or their output that are not already available from GitHub repositories (cited in main text). The scripts are roughly organised by task:

• dataProcessing
• IQTreeOutputProcessing
• IQTreeSearches
• plotting
• renamingScripts
• simulations

From here all subfolders follow these naming conventions:

• Taxon sampling (see main paper):
  • setting6_fast or S6: includes fast-evolving taxa
  • setting7_slow or S7: does not include fast-evolving taxa
• Topology (see Fig. 2 in main paper for tree topologies)
  • MonoDeut: monophyletic Deuterostomia topology
  • ProtAmbu: Orthozoa topology
  • ProtChord: Centroneuralia topology
  • ProtPara: paraphyletic Protostomia
• Amino acid substitution model: LG, CAT, EDM; if ‘_G’ in folder name the models were set up as LG+G, CAT+G or EDM+G (see main paper for methodology details)

alignments.zip folder: contains the concatenated alignments used for all phylogenetic analyses (306-taxon and subsetted alignments)

• concatenated_306taxa: list of taxa and the fasta (.aln) and phylip (.phy) formatted concatenated alignment
• setting6_fast: subsetted alignments including long-branched taxa
  • fasta: folder with 100 fasta-formatted randomly subsetted alignments
  • phylip: folder with 100 phylip-formatted randomly subsetted alignments
• setting7_slow: subsetted alignments excluding long-branched taxa
  • fasta: folder with 100 fasta-formatted randomly subsetted alignments
  • phylip: folder with 100 phylip-formatted randomly subsetted alignments

crossVals.zip: contains the output of leave-one-out cross-validation (loocv) analyses. PhyloBayes run files can be provided on request, omitted from repository to limit size.

• aln68_S7 and aln69_S6: contain the phylip alignments, guide trees and EDM category file used as input to PhyloBayes
• Each output folder contains the *.cpo and *.sitelogl files needed to summarise the lcoov analyses with the PhyloBayes-provided scripts (details pages 16-17 of the manual)

EDCfiles.zip: contains the IQTREE-compatible EDM category file generated with the EDCluster package

inputTrees.zip: contains the guide trees used for the IQTREE topology-scoring analyses

• setting6_fast: contains the pruned trees for 100 subsetted alignments
• setting7_slow: contains the pruned trees for 100 subsetted alignments
• treesToPrune: original 306-taxa unpruned trees for MonoDeut, ProtAmbu and ProtChord hypotheses

IQTreeOutputFiles.zip: contains the output files from IQTREE’s topology-scoring runs, organised by subsetting strategy and substitution model used (e.g. setting6_fast/EDM_G/). Within each of these subfolders analyses are organised into:

• aln#_MonoDeut
• aln#_ProtAmbu
• aln#_ProtChord
• This EXACT folder structure is needed to run the IQTreeOutput_manipulation.R script. Each subfolder contains the standard IQTREE output files, plus a file with the per site log-likelihood score (.sitelh) and a file with the per site rate category (.rate), which are the input files for the IQTreeOutput_manipulation.R script.

results.zip: contains edited files with site likelihoods, rate categories and supported topologies

• pseudoRateCat_noGamma: contains the files needed to generate the supplementary plots S1 and S3
• for each setting*/model/
  • sitelhRatesTables: output of the IQTreeOutput_manipulation.R script, file extensions explained in script.
  • siteSupportedTopology: output of the likelihood_transform.py script. Output of interest compiled in the allReps_labelled.stats file.
• siteProfiles: files needed to generate figure 5 in the main text

topoSims.zip: contains the alignments and the IQTree output for the simulation analyses

• Deut: data simulated under the Deuterostomia topology
  • setting6_fast:
    • deutSup_aln14: data simulated from the long-branched subsetted alignment 14
      • alignments: simulated alignments
      • IQTreeRuns: follow IQTreeOutputFiles.zip structure and naming convention but all analyses ran with the gamma parameter
      • siteProfileSims.zip: PhyloBayes site-profiles and EDCluster EDM category files for simulated alignments 40, 80, 120, 160 and 200
    • paraSup_aln91: data simulated from the long-branched subsetted alignment 91
      • same subfolders as deutSup_aln14
  • setting7_slow:
    • deutSup_aln47: data simulated from the short-branched subsetted alignment 47
      • same subfolders as setting6_fast/deutSup_aln14
    • paraSup_aln62: data simulated from the short-branched subsetted alignment 62
      • same subfolders as setting6_fast/deutSup_aln14
• Orth: data simulated under Orthozoa topology
  • subfolders identical to Deut folder

RELLbootstraps.zip: contains the R script and workspace image to generate the supplementary plots S6-7

Change Log

20 Jun 2025: Minor updates to README.