analyses_metabol_rate_structure/ ├── README.md ├── .gitignore ├── scripts/ # analysis pipeline, run via scripts/run_all.R │ ├── archive/ # superseded scripts kept for reference (not run) │ └── qc_stephan/ # QC scripts for the Stephan volume compilation ├── R/ # shared helpers (plot_settings.R, project_root.R) ├── data_raw/ # source data as published/digitised ├── data_intermediate/ # derived tables passed between scripts ├── data_analysis/ # Study 1b result tables (destination of the s1b_* scripts) ├── metadata/ # provenance, correction log, audit plans, output manifest ├── checks/ # QC / audit outputs (no deliverables) ├── tables/ # deliverable tables, one subfolder per study ├── figs/ # deliverable figures, one subfolder per study ├── network_residual_autocorrelation/ # outputs of the network QC script └── logs/ # pipeline run logs (git-ignored)

Running the pipeline

Rscript scripts/run_all.R

Runs every analysis script in dependency order, then sources scripts/verify_outputs.R as a gate against metadata/output_manifest.csv. See metadata/SESSION_HANDOFF.md for current status and open items.

Data corrections

See metadata/CORRECTIONS_LOG.md for the full log.

• 2026-06-03 — Stephan_primates.csv, Telencephalon, Alouatta seniculus. Value was 37.39 (≈1000× too small). True value from the source table (Stephan et al. 1981) is 37,388; an OCR step misread the thousands-comma as a decimal point (37,388 → 37.388), later rounded to 37.39. Detected because the five fundamental brain parts did not sum to Total_brain_net_volume and the neocortex (NeoWG = 31,660) exceeded the stated telencephalon. Corrected to 37388 in data_raw/Stephan_primates.csv and in all 11 archived copies. Also removed a stray empty column from data_raw/Stephan_primates.csv (now 60 × 69). Re-run derivation scripts so downstream files pick up the fix.