MICOM model databases

Workflows and input data for the construction of model databases for use in MICOM.

See the releases for the Zenodo links to download databases for a specific version.

Local setup

Install Qiime2 following the normal conda instructions. Let's assume your Qiime2 environments is named qiime2-2023.2.

Add the additional requirements.

conda env update -n qiime2-2023.2 -f conda.yml

conda activate qiime2-2023.2

You will now be able to build the databases by running the Nextflow workflows in each directory. If there is a download workflow this has to be run first. For instance, to build AGORA2:

cd recipes/agora2
nextflow run download_agora2.nf -resume
nextflow run agora2.nf -resume

This will create the databases in recipes/agora2/databases.

Organization

root
|---|release_notes
|---|recipes
      |----|[SOURCE]
           |---|[SOURCE].nf         # workflow to generate the DBs
               |data                # required input data
               |manifests           # lists of all models in a DB
               |databases           # databases that were built

Databases

AGORA 1
cite: https://doi.org/10.1038/nbt.3703

Models built from AGORA 1. This was the only available database in prior versions of MICOM.

AGORA 2
cite: https://doi.org/10.1038/s41587-022-01628-0

An updated version of AGORA 1 now containing models for 7,302 strains.

CARVEME
cite: https://doi.org/10.1093/nar/gky537

Automated reconstructions using CARVEME for all bacterial genomes present in Refseq since version 84.

UHGG
cite: https://doi.org/10.1038/s41587-020-0603-3 and https://doi.org/10.1186/s13059-021-02295-1

Automated reconstructions using gapseq for all bacterial and archaeal genomes in UHGG version 2 and gapfilled on the human gut medium.