# ADMET-AI
This git repo contains the code for ADMET-AI, an ADMET prediction platform that
uses [Chemprop-RDKit]((https://github.com/chemprop/chemprop)) models trained on ADMET datasets from the Therapeutics
Data Commons ([TDC](https://tdcommons.ai/)). ADMET-AI can be used to make ADMET predictions on new molecules via the
command line, via the Python API, or via a web server. A live web server hosting ADMET-AI is
at [admet.ai.greenstonebio.com](https://admet.ai.greenstonebio.com)
Please see the following paper and [this blog post](https://portal.valencelabs.com/blogs/post/admet-ai-a-machine-learning-admet-platform-for-evaluation-of-large-scale-QPEa0j5OTYYHTaA) for more
details, and please cite us if ADMET-AI is useful in your work. Instructions to reproduce the results in our paper are in [docs/reproduce.md](docs/reproduce.md).
[ADMET-AI: A machine learning ADMET platform for evaluation of large-scale chemical libraries](https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/btae416/7698030?utm_source=authortollfreelink&utm_campaign=bioinformatics&utm_medium=email&guestAccessKey=f4fca1d2-49ec-4b10-b476-5aea3bf37045)
## Table of Contents
- [Installation](#installation)
- [Predicting ADMET properties](#predicting-admet-properties)
* [Command line tool](#command-line-tool)
* [Python module](#python-module)
* [Web server](#web-server)
## Installation
ADMET-AI can be installed in a few minutes on any operating system using pip (optionally within a conda environment). If
a GPU is available, it will be used by default, but the code can also run on CPUs only.
Optionally, create a conda environment.
```bash
conda create -y -n admet_ai python=3.10
conda activate admet_ai
```
Install ADMET-AI via pip.
```bash
pip install admet-ai
```
Alternatively, clone the repo and install ADMET-AI locally.
```bash
git clone https://github.com/swansonk14/admet_ai.git
cd admet_ai
pip install -e .
```
By default, the pip installation only includes dependencies required for making ADMET predictions, either via the
command line or via the Python API. To install dependencies required for processing TDC data or plotting TDC results,
run `pip install admet-ai[tdc]`. To install dependencies required for hosting the ADMET-AI web server,
run `pip install admet-ai[web]`.
If there are version issues with the required packages, create a conda environment with specific working versions of the
packages as follows.
```bash
pip install -r requirements.txt
pip install -e .
```
Note: If you get the issue `ImportError: libXrender.so.1: cannot open shared object file: No such file or directory`,
run `conda install -c conda-forge xorg-libxrender`.
## Predicting ADMET properties
ADMET-AI can be used to make ADMET predictions in three ways: (1) as a command line tool, (2) as a Python module, or (3)
as a web server.
### Command line tool
ADMET predictions can be made on the command line with the `admet_predict` command, as illustrated below.
```bash
admet_predict \
--data_path data.csv \
--save_path preds.csv \
--smiles_column smiles
```
This command assumes that there exists a file called `data.csv` with SMILES strings in the column `smiles`. The
predictions will be saved to a file called `preds.csv`.
### Python module
ADMET predictions can be made using the `predict` function in the `admet_ai` Python module, as illustrated below.
```python
from admet_ai import ADMETModel
model = ADMETModel()
preds = model.predict(smiles="O(c1ccc(cc1)CCOC)CC(O)CNC(C)C")
```
If a SMILES string is provided, then `preds` is a dictionary mapping property names to values. If a list of SMILES
strings is provided, then `preds` is a Pandas DataFrame where the index is the SMILES and the columns are the
properties.
### Web server
ADMET predictions can be made using the ADMET-AI web server, as illustrated below. Note: Running the following command
requires additional web dependencies (i.e., `pip install admet-ai[web]`).
```bash
admet_web
```
Then navigate to http://127.0.0.1:5000 to view the website.
### Analysis plots
The DrugBank reference plot and radial plots displayed on the ADMET-AI website can be generated locally using the
`scripts/plot_drugbank_reference.py` and `scripts/plot_radial_summaries.py` scripts, respectively. Both scripts
take as input a CSV file with ADMET-AI predictions along with other parameters.
Raw data
{
"_id": null,
"home_page": "https://github.com/swansonk14/admet_ai",
"name": "admet-ai",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.10",
"maintainer_email": null,
"keywords": "machine learning, drug design, ADMET, molecular property prediction",
"author": "Kyle Swanson",
"author_email": "swansonk.14@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/64/86/899ffe85c608d57a68e67840b48487fc7a7831a410d30a2c43ea16ba02d1/admet_ai-1.3.1.tar.gz",
"platform": null,
"description": "# ADMET-AI\n\nThis git repo contains the code for ADMET-AI, an ADMET prediction platform that\nuses [Chemprop-RDKit]((https://github.com/chemprop/chemprop)) models trained on ADMET datasets from the Therapeutics\nData Commons ([TDC](https://tdcommons.ai/)). ADMET-AI can be used to make ADMET predictions on new molecules via the\ncommand line, via the Python API, or via a web server. A live web server hosting ADMET-AI is\nat [admet.ai.greenstonebio.com](https://admet.ai.greenstonebio.com)\n\nPlease see the following paper and [this blog post](https://portal.valencelabs.com/blogs/post/admet-ai-a-machine-learning-admet-platform-for-evaluation-of-large-scale-QPEa0j5OTYYHTaA) for more\ndetails, and please cite us if ADMET-AI is useful in your work. Instructions to reproduce the results in our paper are in [docs/reproduce.md](docs/reproduce.md).\n\n[ADMET-AI: A machine learning ADMET platform for evaluation of large-scale chemical libraries](https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/btae416/7698030?utm_source=authortollfreelink&utm_campaign=bioinformatics&utm_medium=email&guestAccessKey=f4fca1d2-49ec-4b10-b476-5aea3bf37045)\n\n\n## Table of Contents\n\n- [Installation](#installation)\n- [Predicting ADMET properties](#predicting-admet-properties)\n * [Command line tool](#command-line-tool)\n * [Python module](#python-module)\n * [Web server](#web-server)\n\n## Installation\n\nADMET-AI can be installed in a few minutes on any operating system using pip (optionally within a conda environment). If\na GPU is available, it will be used by default, but the code can also run on CPUs only.\n\nOptionally, create a conda environment.\n\n```bash\nconda create -y -n admet_ai python=3.10\nconda activate admet_ai\n```\n\nInstall ADMET-AI via pip.\n\n```bash\npip install admet-ai\n```\n\nAlternatively, clone the repo and install ADMET-AI locally.\n\n```bash\ngit clone https://github.com/swansonk14/admet_ai.git\ncd admet_ai\npip install -e .\n```\n\nBy default, the pip installation only includes dependencies required for making ADMET predictions, either via the\ncommand line or via the Python API. To install dependencies required for processing TDC data or plotting TDC results,\nrun `pip install admet-ai[tdc]`. To install dependencies required for hosting the ADMET-AI web server,\nrun `pip install admet-ai[web]`.\n\nIf there are version issues with the required packages, create a conda environment with specific working versions of the\npackages as follows.\n\n```bash\npip install -r requirements.txt\npip install -e .\n```\n\nNote: If you get the issue `ImportError: libXrender.so.1: cannot open shared object file: No such file or directory`,\nrun `conda install -c conda-forge xorg-libxrender`.\n\n## Predicting ADMET properties\n\nADMET-AI can be used to make ADMET predictions in three ways: (1) as a command line tool, (2) as a Python module, or (3)\nas a web server.\n\n### Command line tool\n\nADMET predictions can be made on the command line with the `admet_predict` command, as illustrated below.\n\n```bash\nadmet_predict \\\n --data_path data.csv \\\n --save_path preds.csv \\\n --smiles_column smiles\n```\n\nThis command assumes that there exists a file called `data.csv` with SMILES strings in the column `smiles`. The\npredictions will be saved to a file called `preds.csv`.\n\n### Python module\n\nADMET predictions can be made using the `predict` function in the `admet_ai` Python module, as illustrated below.\n\n```python\nfrom admet_ai import ADMETModel\n\nmodel = ADMETModel()\npreds = model.predict(smiles=\"O(c1ccc(cc1)CCOC)CC(O)CNC(C)C\")\n```\n\nIf a SMILES string is provided, then `preds` is a dictionary mapping property names to values. If a list of SMILES\nstrings is provided, then `preds` is a Pandas DataFrame where the index is the SMILES and the columns are the\nproperties.\n\n### Web server\n\nADMET predictions can be made using the ADMET-AI web server, as illustrated below. Note: Running the following command\nrequires additional web dependencies (i.e., `pip install admet-ai[web]`).\n\n```bash\nadmet_web\n```\n\nThen navigate to http://127.0.0.1:5000 to view the website.\n\n### Analysis plots\n\nThe DrugBank reference plot and radial plots displayed on the ADMET-AI website can be generated locally using the\n`scripts/plot_drugbank_reference.py` and `scripts/plot_radial_summaries.py` scripts, respectively. Both scripts\ntake as input a CSV file with ADMET-AI predictions along with other parameters.\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "admet_ai",
"version": "1.3.1",
"project_urls": {
"Download": "https://github.com/swansonk14/admet_ai/archive/refs/tags/v_1.3.1.tar.gz",
"Homepage": "https://github.com/swansonk14/admet_ai"
},
"split_keywords": [
"machine learning",
" drug design",
" admet",
" molecular property prediction"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "6486899ffe85c608d57a68e67840b48487fc7a7831a410d30a2c43ea16ba02d1",
"md5": "37a5cfdf8ee584d62541e4861e196550",
"sha256": "7e80c611729853e2b48aa39da5a117978bfd3eef66f94bc2375fa8b2547eee1a"
},
"downloads": -1,
"filename": "admet_ai-1.3.1.tar.gz",
"has_sig": false,
"md5_digest": "37a5cfdf8ee584d62541e4861e196550",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.10",
"size": 16950090,
"upload_time": "2024-07-13T08:44:06",
"upload_time_iso_8601": "2024-07-13T08:44:06.781899Z",
"url": "https://files.pythonhosted.org/packages/64/86/899ffe85c608d57a68e67840b48487fc7a7831a410d30a2c43ea16ba02d1/admet_ai-1.3.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-07-13 08:44:06",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "swansonk14",
"github_project": "admet_ai",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"requirements": [
{
"name": "chemfunc",
"specs": [
[
"==",
"1.0.4"
]
]
},
{
"name": "chemprop",
"specs": [
[
"==",
"1.6.1"
]
]
},
{
"name": "flask",
"specs": [
[
"==",
"3.0.0"
]
]
},
{
"name": "gunicorn",
"specs": [
[
"==",
"21.2.0"
]
]
},
{
"name": "numpy",
"specs": [
[
"==",
"1.25.2"
]
]
},
{
"name": "openpyxl",
"specs": [
[
"==",
"3.1.2"
]
]
},
{
"name": "pandas",
"specs": [
[
"==",
"2.1.0"
]
]
},
{
"name": "PyTDC",
"specs": [
[
"==",
"0.4.1"
]
]
},
{
"name": "rdkit",
"specs": [
[
"==",
"2023.3.3"
]
]
},
{
"name": "seaborn",
"specs": [
[
"==",
"0.13.0"
]
]
},
{
"name": "tqdm",
"specs": [
[
"==",
"4.66.1"
]
]
},
{
"name": "typed-argument-parser",
"specs": [
[
"==",
"1.9.0"
]
]
}
],
"lcname": "admet-ai"
}
JSON
