Installation#

Hint

Windows OS is not supported. Consider using a Python inside Windows Subsystem for Linux.

Requirements#

  • Python: >= 3.9, < 3.13 (tested with 3.11)

  • Dependencies: Automatically installed via pip

Installation from PyPI#

Install using pip (recommended):

pip install mdeepfri

Installation from Source#

For development or the latest features:

git clone https://github.com/bioinf-mcb/Metagenomic-DeepFRI.git
cd Metagenomic-DeepFRI
pip install -e .

Download Models#

Two versions of models are available:

  • v1.0: Original version from DeepFRI publication

  • v1.1: Finetuned on AlphaFold models and machine-generated Gene Ontology Uniprot annotations

To download models run command:

mDeepFRI get-models --output path/to/weights/folder -v {1.0 or 1.1}

Prepare Structural Database#

The PDB database will be automatically downloaded and installed during the first run of mDeepFRI.

You can download additional FoldComp databases from the FoldComp website. During first run, FASTA sequences will be extracted from FoldComp database and MMseqs2 database will be created and indexed.

Tested databases:

  • afdb_swissprot

  • afdb_swissprot_v4

  • afdb_rep_v4

  • afdb_rep_dark_v4

  • afdb_uniprot_v4

  • esmatlas

  • esmatlas_v2023_02

  • highquality_clust30

Warning

Do not rename downloaded databases. FoldComp has certain inconsistencies in the way FASTA sequences are extracted, therefore the pipeline was tweaked for each database.

For custom FoldComp databases, use the FoldComp executable:

foldcomp compress [-t number] <dir|tar(.gz)> [<dir|tar|db>]

Verification#

Verify installation:

mDeepFRI --help
python -c "import mDeepFRI; print(mDeepFRI.__version__)"