• Stars
    star
    198
  • Rank 195,693 (Top 4 %)
  • Language
    Python
  • License
    GNU General Publi...
  • Created about 4 years ago
  • Updated 23 days ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

The Ersilia Model Hub, a repository of AI/ML models for infectious and neglected disease research.

Welcome to the Ersilia Model Hub

Donate Contributor Covenant License: GPL v3 DOI

documentation PyPI version fury.io Python 3.7 Code style: black

Table of Contents

  1. Project Description
  2. Quick start guide
  3. Contribute
  4. License and citation
  5. About us

Project Description

The Ersilia Model Hub is a unified platform of pre-trained AI/ML models for infectious and neglected disease research. The end goal is to provide an open-source, low-code solution to access AI/ML models for drug discovery. The models embedded in the hub include both models published in the literature (with appropriate third party acknowledgement) and models developed by the Ersilia team or contributors.

Quick Start Guide

Please check the package requirements in the Installation Guide. The next steps are a quickstart guide to installing Ersilia.

  1. Create a conda environment and activate it

    conda create -n ersilia python=3.10
    conda activate ersilia
  2. Clone this repository and install with pip

    git clone https://github.com/ersilia-os/ersilia.git
    cd ersilia
    pip install -e .
  3. Once the Ersilia Model Hub is installed, you can use the CLI to run predictions. First, select a model from the Ersilia Model Hub and fetch it:

    ersilia fetch retrosynthetic-accessibility
  4. Generate a few (5) example molecules, to be used as input. The example command will generate the adequate input for the model in use

    ersilia example retrosynthetic-accessibility -n 5 -f my_molecules.csv
  5. Then, serve your model:

    ersilia serve retrosynthetic-accessibility
  6. And run the model:

    ersilia run -i my_molecules.csv -o my_predictions.csv
  7. Finally, close the service when you are done.

    ersilia close
  8. If you no longer want to use the model, you can delete it.

ersilia delete retrosynthetic-accessibility

Please see the Ersilia Book for more examples and detailed explanations.

Contribute

The Ersilia Model Hub is a Free, Open Source Software and we highly value new contributors. There are several ways in which you can contribute to the project:

  • A good place to start is checking open issues.
  • If you have identified a bug in the code, please open a new issue using the bug template.
  • Share any feedback with the community using GitHub Discussions for the project
  • Check our Contributing Guide for more details

The Ersilia Open Source Initiative adheres to the Contributor Covenant code of conduct.

Submit a New Model

If you want to incorporate a new model in the platform, open a new issue using the model request template or contact us using the following form.

After submitting your model request via an issue (suggested), a maintainer will review your request. If they /approve your request, a new model respository will be created for you to fork and use! There is a demo repository explaining the steps one-by-one.

License and Citation

This repository is open-sourced under the GPL-3 License. Please cite us if you use it!

About Us

The Ersilia Open Source Initiative is a Non Profit Organization (1192266) with the mission is to equip labs, universities and clinics in LMIC with AI/ML tools for infectious disease research.

Help us achieve our mission or volunteer with us!

More Repositories

1

chempfn

Ensemble-based, size-agnostic wrapper for the TabPFN classifier
Jupyter Notebook
28
star
2

zaira-chem

Automated QSAR based on multiple small molecule descriptors
Python
27
star
3

chem-sampler

Sample small molecules, both from large chemical libraries and generated
Python
10
star
4

chemxor

Privacy preserving machine learning for small molecule data
Python
9
star
5

event-fund-ai-drug-discovery

Coding and data materials for the Event Fund AI for Drug Discovery Course
Jupyter Notebook
7
star
6

eos2v11

ADMETlab 2 for systematic evaluation of ADMET properties
Python
6
star
7

compound-embedding

Compound embedding obtained with few-shot learning (FS-Mol), Morgan fingerprints, Grover embeddings, and Mordred descriptors
Jupyter Notebook
6
star
8

olinda

Chemistry model distillation, based on 1024-dimensional embeddings
Python
4
star
9

osm-series4-candidates

New candidates for the Open Source Malaria series 4 compounds
Python
4
star
10

eos8a4x

RDKIT physicochemical descriptors of small molecules
Python
3
star
11

eos7asg

PADEL physicochemical descriptors
Python
3
star
12

eos78ao

Mordred physicochemical descriptors
Python
3
star
13

osa-murd

Open Source Antibiotics MurD ligands based on generative models.
Jupyter Notebook
3
star
14

eos1af5

Explainable AI for Caco-2 cell wall permeability
Jupyter Notebook
3
star
15

pharmacogx-embeddings

Pharmacogenomics knowledge graph embeddings and related analyses
Jupyter Notebook
3
star
16

eos5axz

Morgan count fingerprints with radius 3
Python
2
star
17

eos7cpf

Predicts the Lipophilicity (octanol/water distribution coefficient) of small molecules
Python
2
star
18

python101

Materials for the Python Introductory Course
Jupyter Notebook
2
star
19

eos7pw8

Python
2
star
20

eos9ei3

Estimation of synthetic accessibility score of drug-like molecules based on molecular complexity and fragment contributions
Python
2
star
21

eos3ae6

Holistic molecular descriptors for scaffold hopping
Python
2
star
22

eos3ae7

Chemical Variational Autoencoder (VAE)
Python
2
star
23

osm-series4-candidates-2

A new round of series 4 candidates, including new low-data generative models and improved predictors
Python
2
star
24

cidrz-e2e-linkage

A fully automated pipeline for fuzzy matching of medical datasets.
Python
2
star
25

eos481p

grover-toxcast
Python
2
star
26

eos2re5

A platform for systematic ADME evaluation of drug molecules
Python
2
star
27

eos2ta5

Ligand-based prediction of hERG blockade
Python
2
star
28

stylia

Decent scientific plot styles
Jupyter Notebook
2
star
29

eos1amr

Predictor of blood-brain-barrier permeability
Python
2
star
30

eos54ij

Cardiac Toxicity Prediction
Python
2
star
31

eos1579

Classification of carcinogenic metabolites
Python
2
star
32

eos5smc

grover-tox21
Python
2
star
33

eos2lm8

SMILES transformer embeddings
Python
2
star
34

griddify

Griddify high-dimensional tabular data for easy visualization and deep learning
Jupyter Notebook
2
star
35

antimicrobial-ml-tasks

Antimicrobial activity prediction with automated machine learning
Jupyter Notebook
2
star
36

eos43at

Explainable AI applied to hERG blockade
Python
2
star
37

model-inference-pipeline

Ersilia's batch inference pipeline on the AWS cloud
Jupyter Notebook
2
star
38

eos1xje

Python
2
star
39

open-data-cleaning

Repository with notebooks that clean open data files downloaded from different sources
Jupyter Notebook
1
star
40

eos6ao8

Explainable AI for plasma protein binding prediction
Jupyter Notebook
1
star
41

lazy-qsar

A library to build fast QSAR models
Python
1
star
42

eos8lok

Python
1
star
43

eos7w6n

Grover embedding for molecular representation
Python
1
star
44

isaura

A lake of precalculated properties of biomedical entities based on the Ersilia Model Hub
Python
1
star
45

eos2mrz

DeepSMILES
Python
1
star
46

PyBioMed

Python
1
star
47

mock-website

SCSS
1
star
48

eos8ioa

A simple natural product likeness score
Python
1
star
49

eos92sw

Predicts the toxicity and synthetic availability of small molecules
Python
1
star
50

eos2b6f

Jupyter Notebook
1
star
51

eos-template

Template repository to add new models to the Ersilia Model Hub
Python
1
star
52

eos8bhe

Python
1
star
53

eos2r5a

Retrosynthetic accessibility score
Python
1
star
54

eos7ack

SwissADME
Python
1
star
55

chemical-library-exploration

This simple repository performs a fast exploration of a chemical library
Python
1
star
56

osm-series4-synthesis-round1

Selection of candidates for first round of experimental synthesis
HTML
1
star
57

bioassay-db

Scripts to download data from ChEMBL
Python
1
star
58

ersilia-book

Ersilia Book repository
1
star
59

eos8451

grover-esol
Python
1
star
60

eos4u6p

Chemical Checker 25 bioactivity signatures
Python
1
star
61

easylinkage

Standard and extended record linkage functionalities
Python
1
star
62

zaira-chem-tdc

This repository contains the benchmarking of ZairaChem using the Therapeutics Data Commons Datasets
Jupyter Notebook
1
star
63

eos31ve

Python
1
star
64

eos69p9

Chemical toxicity prediction based on SSL-GCN
Python
1
star
65

eos238c

mesh-therapeutic-use
Python
1
star
66

eos7a04

Continuous and data driven descriptors
Python
1
star
67

auto-tabnet

TabNet implementation with hyperparameter search
Python
1
star
68

autocat-chem

Automatic training of (multioutput) regression models for chemistry
Python
1
star
69

eos8d8a

Mycobacterium tuberculosis membrane permeability of drugs
Perl
1
star
70

eos526j

A tool for planning retrosynthesis of a target molecule based on template reactions and a stock of precursors
Python
1
star
71

pharmacogx-arsa

Look for abundant region specific alleles for pharmacogenomics
Jupyter Notebook
1
star
72

pabb-docking

Docking experiments for the pabb protein
Jupyter Notebook
1
star
73

eos157v

grover-freesolv
Python
1
star
74

eos6fza

grover-clintox
Python
1
star
75

fps-grid

Converts molecular fingerprints to grids for image-based DL
Jupyter Notebook
1
star
76

eos7jio

RDKIT path-based fingerprints
Python
1
star
77

eos2thm

MolBERT molecular representation
Python
1
star
78

eos9be7

Frechet Chemnet Distance
Python
1
star
79

groverfeat

An interface to the GROVER model for molecule featurization
Python
1
star
80

osm-pfatp4-structure

Jupyter Notebook
1
star
81

eos6oli

Aqueous solubility prediction
Python
1
star
82

eos93ox

Classification of hERG blockers and nonblockers
Python
1
star
83

eos2fy6

Python
1
star
84

eos4cxk

Python
1
star
85

eos3d3r

CardPred hERG blockade
Python
1
star
86

eos8h6g

Avalon fingerprints
Python
1
star
87

eos30gr

Python
1
star
88

eos2lqb

Python
1
star
89

eos77w8

Predicts side-effects of small molecule drugs
Python
1
star
90

h3d-screening-cascade-code

Code for the manuscript entitled "First fully automated AI/ML virtual screening cascade implemented at a drug discovery centre in Africa"
Python
1
star
91

ai2050-h3d-symposium-workshop

Material for a 2h introductory workshop on AI/ML for drug discovery
Jupyter Notebook
1
star