Improving the generalizability of protein-ligand binding predictions with AI-Bind

Release for publication in Nature Communications. This release updates https://github.com/Barabasi-Lab/AI-Bind/Emergence-of-shortcuts/ from v1.0 release, adding in code for the 3 scenarios of the toy model described in the Supplementary Information of the paper.

AI-Bind_v1.0

First release of AI-Bind as described by "AI-Bind: Improving Binding Predictions for Novel Protein Targets and Ligands" on arXiv. For citation purposes:

@misc{chatterjee2021aibind,
doi = {10.48550/ARXIV.2112.13168},
url = {https://arxiv.org/abs/2112.13168},
author = {Chatterjee, Ayan and Walters, Robin and Shafi, Zohair and Ahmed, Omair Shafi and Sebek, Michael and Gysi, Deisy and Yu, Rose and Eliassi-Rad, Tina and Barabási, Albert-László and Menichetti, Giulia},
keywords = {Quantitative Methods (q-bio.QM), Machine Learning (cs.LG), FOS: Biological sciences, FOS: Biological sciences, FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {AI-Bind: Improving Binding Predictions for Novel Protein Targets and Ligands},
publisher = {arXiv},
year = {2021},
copyright = {arXiv.org perpetual, non-exclusive license}
}

Soon to be published in Nature Communications.