Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
EquiBind: geometric deep learning for fast predictions of the 3D structure in which a small molecule binds to a protein
| Date | Stars |
|---|---|
| 2026-07-31 | 547 |
| 2026-08-05 | 547 |
| 2026-08-06 | 547 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction ### [Paper on arXiv](https://arxiv.org/abs/2202.05146) **Before using EquiBind, also consider checking out our new approach called DiffDock which improves over EquiBind in multiple ways. The DiffDock [GitHub](https://github.com/gcorso/DiffDock) and [paper](https://arxiv.org/abs/2210.01776).** EquiBind, is a SE(3)-equivariant geometric deep learning model performing direct-shot prediction of both i) the receptor binding location (blind docking) and ii) the ligand’s bound pose and orientation. EquiBind achieves significant speed-ups compared to traditional and recent baselines. If you have questions, don't hesitate to open an issue or ask me via [[email protected]]([email protected]) or [social media](https://hannes-stark.com/) or Octavian Ganea via [[email protected]]([email protected]). We are happy to hear from you!   # Dataset Our preprocessed data (see dataset section in the paper Appendix) is no longer available on [zenodo](https://zenodo.org/record/6408497). \ The reason is that the PDBBind license does not allow for redistributing the dataset. The files in `data` contain the names for the time-based data split. This means that now you have to download the data from the PDBBind website (http://www.pdbbind.org.cn/) and place it into `data` such that you have the path `data/PDBBind` # Use provided model weights to predict binding structure of your own protein-ligand pairs: ## Step 1: What you need as input Ligand files of the formats ``.mol2`` or ``.sdf`` or ``.pdbqt`` or ``.pdb`` whose names contain the string `ligand` (your ligand files should contain **all** hydrogens). \ Receptor files of the format ``.pdb`` whose names contain the string `protein`. We ran [reduce](https://github.com/rlabduke/reduce) on our training proteins. Maybe you also want to run it on your protein.\ For each complex you want to predict you need a directory containing the ligand and receptor file. Like this: ``` my_data_folder └───name1 │ name1_protein.pdb │ name1_ligand.sdf └───name2 │ name2_protein.pdb │ name2_ligand.mol2 ... ``` ## Step 2: Setup Environment We will set up the environment using [Anaconda](https://docs.anaconda.com/anaconda/install/index.html). Clone the current repo git clone https://github.com/HannesStark/EquiBind Create a new environment with all required packages using `environment.yml`. If you have a CUDA GPU run: conda env create -f environment.yml If you instead only have a CPU run: conda env create -f environment_cpuonly.yml Activate the environment conda activate equibind Here are the requirements themselves for the case with a CUDA GPU if you want to install them manually instead of using the `environment.yml`: ```` python=3.7 pytorch 1.10 torchvision cudatoolkit=10.2 torchaudio dgl-cuda10.2 rdkit openbabel biopython rdkit biopandas pot dgllife joblib pyaml icecream matplotlib tensorboard ```` ## Step 3: Predict Binding Structures! In the config file `configs_clean/inference.yml` set the path to your input data folder `inference_path: path_to/my_data_folder`. Then run: python inference.py --config=configs_clean/inference.yml Done! :tada: \ Your results are saved as `.sdf` files in the directory specified in the config file under ``output_directory: 'data/results/output'`` and as tensors at ``runs/flexible_self_docking/predictions_RDKitFalse.pt``! # Inference for multiple ligands in the same .sdf file and a single receptor python multiligand_infernce.py -o path/to/output_directory -r path/to/receptor.pdb -l path/to/ligands.sdf This runs EquiBind on every ligand in ligands.sdf against the protein in receptor.pdb. The outputs are 3 files in output_directory with the following names and contents: failed.txt - contains the index (in the file ligands.sdf) and name of every molecule for which inference failed in a way that was caught and handled.\ success
Excerpt of 5,791 characters
Read on GitHub
31
6
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:22bcd47687297f7c, topic:drug-discovery, desc:protein, readme:protein