Genome-wide strengthening of evolutionary constraint across the volvocine multicellularity gradient — data, code, and supplementary tables

This Zenodo record contains the full analysis pipeline and reproducibilitypackage accompanying the manuscript: "Genome-wide strengthening of evolutionary constraint across the volvocinemulticellularity gradient" by Masato Tanigawa (submitted to Journal ofMolecular Evolution, 2026). Contents (79 MB tarball, 68 files after unpacking): * scripts/ Analysis pipeline (Python + R + bash), all paths auto-resolve from PROJECT_ROOT (no hardcoded paths). * input_data/ OrthoFinder single-copy orthogroups (1,755 × 10 species), supermatrix-derived species tree, Augustus predictions. * intermediate_data/ Per-orthogroup PAML m2/m1 and IQ-TREE outputs (compressed tar.gz, 61 MB combined). * summary_tables/ Headline CSVs including PAML branch-model summaries, IQ-TREE terminal branch lengths, per-species PGLS fit, COG breakdown, and boundary-condition table. * figures/ Figures 1–3 and Supplementary Figures S1–S2 (PDF + PNG). * supplementary/ Supplementary Tables S1, S2, S3, S4, S4b, S5 and full reference database (references.bib + .md). * requirements.txt Python package versions used. * sessionInfo.txt R session info used. * SHA256SUMS Checksums for verification. * LICENSE-DATA CC BY 4.0 (data, figures, tables). * LICENSE-CODE MIT (scripts). Headline results: * Branch-model d_N/d_S: 88.6 % of 845 post-QC orthogroups have ω_fg < ω_bg (multicellular vs Chlamydomonas). * IQ-TREE amino-acid branch length: 96.1 % (1,395/1,451) of complete-data orthogroups show negative Spearman correlation with log2(cell number); species-label permutation test empirical one-sided P = 0.0107 (10,000 permutations). * Per-species PGLS regression (n = 10): β = −0.123, P = 0.002, R² = 0.72, Pagel's λ_ML ≈ 0. * Signal distributed across all 20 COG functional categories analysed (range 78–100 % per category). The main manuscript text is not included in this record; a preprint isdeposited separately on bioRxiv.