Home

Summary

ASMC combines (i) homology modeling of family members (MODELLER), (ii) ligand-binding pocket search (P2RANK), (iii) structural alignment of modeled active sites (USalign) and (iv) density-based spatial clustering of obtained alignments (DBSCAN) in a single command line. Clustering step can be carried out on either structural or sequence alignment.

ASMC is entirely written in Python. At the root of the project, the `run_asmc.py' script is used to run the pipeline.

Installation

Download

Download the latest GitHub release to obtained the code: https://github.com/labgem/ASMC/releases

Python requirements

• Python>=3.8
• numpy
• scikit-learn
• pyyaml
• pillow
• biopython>=1.81
• weblogo

You can install the python dependencies with pip, conda or mamba with theses command and the files given in the releases:

pip

pip install -r requirements.txt

conda

conda env create -n env_name -f env.yml

mamba

mamba env create -n env_name -f env.yml

Installation via conda and mamba includes the modeller installation, but you still need to request the licence key.

External Software dependencies

• P2RANK - for ligand-binding pocket detection (https://github.com/rdk/p2rank)
• MODELLER - for homology modeling (https://salilab.org/modeller/)
• USalign - for structural alignment (https://zhanggroup.org/US-align/)

Please follow the links above and the instructions given by their authors.

Configuration

In ASMC/ressources, add a file exactly named config_asmc.yml. This file should contain 3 informations:

• the path to the ASMC/ressources/AA_distances.tsv
• the path or alias of P2RANK executable (or binary name if it's in your PATH)
• the path or alias of USalign executable (or binary name if it's in your PATH

Example:

distances: "/home/User/ASMC/ressources/AA_distances.tsv"
usalign: "USalign"
p2rank: "prank"

The keys should be identical to this example