Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/biomedsciai/fuse-drug

FuseMedML based molecular biochemistry library for drug discovery/repurposing

https://github.com/biomedsciai/fuse-drug

Science Score: 26.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (12.0%) to scientific vocabulary
Last synced: 10 months ago · JSON representation

Repository

FuseMedML based molecular biochemistry library for drug discovery/repurposing

Basic Info
  • Host: GitHub
  • Owner: BiomedSciAI
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Size: 5.08 MB
Statistics
  • Stars: 22
  • Watchers: 6
  • Forks: 5
  • Open Issues: 12
  • Releases: 0
Created over 3 years ago · Last pushed over 1 year ago
Metadata Files
Readme License

README.md

fuse-drug

FuseMedML based molecular biochemistry library for drug discovery/repurposing

fuse-drug contains generic tools to facilitate working with datasets and representations for proteins, small molecules, and interactions. These include data loaders, processing and augmentation fuse style ops, utilities and more.

It also contains end to end examples of data and model training pipelines, currently focused on the protein-ligand affinity prediction task.

fuse-drug is a work in progress. It will gradually expand to cover more representations and tasks such as molecule property prediction, protein-protein interaction, generation and more. It will also extend the fuse.eval package to cover evaluation metrics specific to the biochemistry domain.

Coming soon: DrugDiscoveryFoundationBenchmarks - A repository for biochemical ML benchmarks, which includes tools for data curation and creation, creating different types of splits for model training, and application of evaluation metrics.

Installation instructions

  1. Install FuseMedML and its dependencies as described here.

  2. Install Fuse-Drug only (without examples) by running: ``` pip install -e .

to also install development deps use:

pip install -e .[dev] ``` or:

Install Fuse-Drug with examples by running: pip install -e .[examples]

In case of a CUDA related error, we recommend working in a conda environment with Python>=3.9 (Create one by running conda create -n ENV_NAME python=3.9) and updating PyTorch following the official PyTorch installation instructions after completing the above steps.

  1. [Optional] Install abnumber python package. This package is used for antibody numbering and alignment (see antibody.py) conda install -c bioconda abnumber

Owner

  • Name: BiomedSciAI
  • Login: BiomedSciAI
  • Kind: organization

GitHub Events

Total
  • Watch event: 7
  • Delete event: 4
  • Member event: 2
  • Push event: 55
  • Pull request event: 31
  • Pull request review event: 33
  • Pull request review comment event: 28
  • Create event: 15
Last Year
  • Watch event: 7
  • Delete event: 4
  • Member event: 2
  • Push event: 55
  • Pull request event: 31
  • Pull request review event: 33
  • Pull request review comment event: 28
  • Create event: 15

Issues and Pull Requests

Last synced: 10 months ago

All Time
  • Total issues: 15
  • Total pull requests: 101
  • Average time to close issues: about 1 month
  • Average time to close pull requests: 7 days
  • Total issue authors: 4
  • Total pull request authors: 11
  • Average comments per issue: 0.4
  • Average comments per pull request: 0.25
  • Merged pull requests: 89
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 0
  • Pull requests: 17
  • Average time to close issues: N/A
  • Average time to close pull requests: 8 days
  • Issue authors: 0
  • Pull request authors: 5
  • Average comments per issue: 0
  • Average comments per pull request: 0.0
  • Merged pull requests: 15
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
  • SagiPolaczek (11)
  • alex-golts (3)
  • mosheraboh (2)
  • sivanravidos (2)
  • floccinauc (1)
  • liamhazan (1)
Pull Request Authors
  • floccinauc (31)
  • YoelShoshan (28)
  • SagiPolaczek (24)
  • mosheraboh (19)
  • alex-golts (15)
  • matanninio (10)
  • michalozeryflato (6)
  • liamhazan (4)
  • simona-rc (1)
  • itaijj (1)
  • bensha6757 (1)
  • shatz01 (1)
  • Sai-Suraj-27 (1)
  • edenjenzohar (1)
  • IdoAmosIBM (1)
Top Labels
Issue Labels
bug (2) good first issue (1) enhancement (1)
Pull Request Labels
enhancement (1)

Dependencies

.github/workflows/lint.yaml actions
  • actions/checkout v2 composite
  • actions/setup-python v2 composite
  • psf/black stable composite
  • py-actions/flake8 v2 composite
fusedrug_examples/requirements.txt pypi
  • biopython *
  • deepchem *
  • dgl *
  • dscript *
  • modlamp *
  • pysmiles *
  • tensorflow *
  • torchtext ==0.3.1
  • x_transformers *
requirements/requirements.txt pypi
  • DeepPurpose *
  • PyTDC *
  • biopython *
  • click *
  • dm-tree *
  • hydra-core *
  • ml_collections *
  • pandarallel *
  • pg8000 *
  • psycopg2-binary *
  • pyfastx *
  • pytorch-lightning <2.0.0
  • rdflib *
  • rdflib-sqlalchemy *
  • rdkit *
  • sqlalchemy *
  • termcolor *
  • tokenizers *
  • torch *
  • torchvision *
  • transformers *
  • xmlrunner *
requirements/requirements_dev.txt pypi
  • black ==22.3.0 development
  • flake8 * development
  • ipykernel * development
  • jupyter * development
  • mypy ==0.950 development
  • nb-clean * development
  • pre-commit * development
setup.py pypi