MACADAM

Coming soon

The runner isn't wired yet. Module slot is reserved in the Microbiome portal so the left-nav shape stays stable once it ships.

MACADAM ("MetAboliC pAthways DAtabase for Microorganisms") rolls a per-sample taxonomic abundance table into per-pathway completeness scores by intersecting each observed taxon with pre-computed MetaCyc, KEGG, and FAPROTAX pathway annotations for the matched reference genome.

Comparator positioning

MACADAM sits next to HUMAnN3 as a functional-prediction comparator, but:

• HUMAnN3 uses read depth → gene → pathway abundance.
• MACADAM uses taxonomic abundance × per-genome pathway completeness.
• GeMMA uses taxonomic abundance × genome-scale metabolic model × metabolomics.

Using all three on the same cohort gives the paper three independent functional readouts from increasingly model-heavy assumptions.

Required inputs

• Per-sample taxonomic abundance table (already on every preprocessed metagenomics dataset).
• MACADAM reference release — pinned to a semver in the same manifest pattern as vmh_metabolites.manifest.json, downloaded lazily to $MACADAM_DB.

Planned outputs

• Per-sample pathway-completeness matrix.
• Per-contrast Wilcoxon differential (MACADAM completeness A vs B).
• Shared-axis panel against HUMAnN3 + GeMMA enrichment so pathway-ID overlap (where the vocabularies intersect) is visible.