dram

name	dram
description	Distill microbial metabolism annotations for MAGs/genomes into curated product tables (DRAM annotate + distill). Use for ecological functional potential, pathway summaries, and MAG metabolism figures. Upstream: https://github.com/WrightonLabCSU/DRAM.
license	MIT
category	analysis-tools
tags	["metabolism","annotation","genome-mining","DRAM"]
upstream	https://github.com/WrightonLabCSU/DRAM
stage	mining

DRAM

Upstream: WrightonLabCSU/DRAM

Citation

Shaffer, M. et al. DRAM for distilling microbial metabolism to automate the curation of microbiome function. Nucleic Acids Res. 48, 8883–8900 (2020). https://doi.org/10.1093/nar/gkaa621

Analytical thinking

DRAM wraps multiple DBs (KEGG, MEROPS, CAZy, … depending on install) then distills annotations into interpretable metabolism sheets. Ideal after binning QC when the question is “what can these MAGs do?”, not read-level HUMAnN pathways.

Heavy one-time DB install — share a read-only DRAM DB on the cluster. Pin DRAM version + DB pull date. Low-completeness MAGs yield sparse distillates; filter with mag-qc first.

Complementary: eggnog-mapper (orthology), antismash (BGCs), gapseq (GEMs). Do not merge DRAM distillate rows with HUMAnN pathway abundances without stating different evidence layers.

Prerequisites

Requirement	Minimum	Recommended
DRAM DB	Configured via DRAM manuals	Shared $DRAM_DATA
Input	Genome/MAG FASTA directory	CheckM2-filtered
RAM	64 GB	128 GB+ for large batches
CPU	8	16–32

How to run

# After DRAM + databases are installed/configured (see upstream docs):
mkdir -p temp/dram
DRAM.py annotate \
  -i 'bins_hq/*.fa' \
  -o temp/dram/annotate \
  --threads 16

DRAM.py distill \
  -i temp/dram/annotate/annotations.tsv \
  -o temp/dram/distill \
  --trna_path temp/dram/annotate/trnas.tsv \
  --rrna_path temp/dram/annotate/rrnas.tsv

Exact flag names can vary slightly by DRAM version — verify with DRAM.py annotate -h / distill -h.

Key products

Output	Use
annotations.tsv	Per-gene annotations
distillate / product sheets	Metabolism summary heatmaps
genome stats	Completeness-aware interpretation

Decision tree

Functional question?
├─ Read-level pathways → humann
├─ MAG metabolism distillate → dram
├─ Genome-scale model → gapseq (HQ MAGs)
└─ BGCs only → antismash

Troubleshooting

Symptom	Action
DB not configured	Complete DRAM database setup before annotate
Annotate dies on one bad FASTA	Validate FASTA headers/unique names
Empty distillate	Check annotate finished; tRNA/rRNA paths

Related skills

microbial-mining · eggnog-mapper · bakta · gapseq · antismash · dbcan · mag-qc