Here you will find basic instructions for running DRAM2, but for full documentation, please see the official DRAM2 webpage: Read-the-docs
DRAM1 can currently still be found in the master branch for now.
DRAM2 (Distilling and Refining Annotations of Metabolism, version 2) is a tool for annotating genomic and metagenomic assemblies (e.g., scaffolds or contigs) as well as predicted genes (nucleotide or amino acid sequences). It organizes genome annotations into metabolic functions across three levels of increasing interpretation: (1) ANNOTATE, (2) SUMMARIZE, and (3) VISUALIZE. This workflow enables the analysis of large numbers of microbial genomes or metagenomes, highlighting functional guilds and supporting inference of organismal metabolism across datasets.
During the ANNOTATE stage, DRAM2 identifies genes in input sequences and annotates them using multiple databases, including KEGG (if provided by the user), UniRef90, PFAM, dbCAN3, RefSeq Viral, VOGDB, MEROPS, and optional user-defined databases. A full list of available annotation databases can be found here: BortonWrightonLabs/dram pipeline parameters. ANNOTATE then integrates results across all databases, increasing annotation coverage and yielding ~25% more database hits than commonly used annotators such as DFAST, MetaERG, and Prokka.
The ANNOTATE output contains all database hits for every gene in each genome, generating a comprehensive output of most annotation pipelines. DRAM2 extends beyond this by organizing (SUMMARIZE) and visualizing (VISUALIZE) annotations into ecosystem-relevant functional categories, enabling more interpretable comparisons across genomes and ecosystems.
Below is an example of basic DRAM2 usage. This code is for annotating a directory of genomes, renaming them for downstream use, calling genes and annotating them using all available databases, performing quality control, summarizing and visualizing with particular ecosystems in mind and assigning genome-level traits to the organisms. The command is submitted on the command line and will run in the background.
nextflow run BortonWrightonLabs/DRAM -r dev --input_fasta [INPUT_FASTA] --outdir [OUTPUT_DIR] --rename --annotate --anno_dbs all --qc --summarize --sum_ecos 'eng_sys,ag' --visualize --traits -profile singularity,slurm -c cluster.config -resume -bg--input_fasta is a directory of assembly FASTAs. These DRAM2 examples use the dev branch; pin a DRAM2 tag or commit for reproducible analyses. --anno_dbs all enables databases whose configured paths exist and skips missing paths with a warning.
Prepare cluster.config for your SLURM partition/account and configure database paths before running. Nextflow options use one dash (-c, -profile, -params-file), while DRAM parameters use two (--input_fasta). See the installation guide for SLURM setup and the usage guide for custom configs, YAML/JSON parameters files, and resource-label overrides. Keep the Nextflow driver running according to your cluster's policy; -bg does not submit the driver to SLURM.
If DRAM helps you in your research, please cite: DRAM publication in Nucleic Acids Research (2020)