Knowledge Enhanced Representation Learning for Drug Discovery
Thanh Lam Hoang, Marco Luca Sbodio, Marcos Martinez Galindo, Mykhaylo Zayats, Raul Fernandez-Diaz, Victor Valls, Gabriele Picco, Cesar Berrospi
Abstract
Recent research on predicting the binding affinity between drug molecules and proteins use representations learned, through unsupervised learning techniques, from large databases of molecule SMILES and protein sequences. While these representations have significantly enhanced the predictions, they are usually based on a limited set of modalities, and they do not exploit available knowledge about existing relations among molecules and proteins. Our study reveals that enhanced representations, derived from multimodal knowledge graphs describing relations among molecules and proteins, lead to state-of-the-art results in well-established benchmarks (first place in the leaderboard for Therapeutics Data Commons benchmark ``Drug-Target Interaction Domain Generalization Benchmark", with an improvement of 8 points with respect to previous best result). Moreover, our results significantly surpass those achieved in standard benchmarks by using conventional pre-trained representations that rely only on sequence or SMILES data. We release our multimodal knowledge graphs, integrating data from seven public data sources, and which contain over 30 million triples. Pretrained models from our proposed graphs and benchmark task source code are also released.
BibTeX
@article{Hoang_Sbodio_Martinez Galindo_Zayats_Fernandez-Diaz_Valls_Picco_Berrospi_Lopez_2024, title={Knowledge Enhanced Representation Learning for Drug Discovery}, volume={38}, url={https://ojs.aaai.org/index.php/AAAI/article/view/28924}, DOI={10.1609/aaai.v38i9.28924}, abstractNote={Recent research on predicting the binding affinity between drug molecules and proteins use representations learned, through unsupervised learning techniques, from large databases of molecule SMILES and protein sequences. While these representations have significantly enhanced the predictions, they are usually based on a limited set of modalities, and they do not exploit available knowledge about existing relations among molecules and proteins. Our study reveals that enhanced representations, derived from multimodal knowledge graphs describing relations among molecules and proteins, lead to state-of-the-art results in well-established benchmarks (first place in the leaderboard for Therapeutics Data Commons benchmark ``Drug-Target Interaction Domain Generalization Benchmark", with an improvement of 8 points with respect to previous best result). Moreover, our results significantly surpass those achieved in standard benchmarks by using conventional pre-trained representations that rely only on sequence or SMILES data. We release our multimodal knowledge graphs, integrating data from seven public data sources, and which contain over 30 million triples. Pretrained models from our proposed graphs and benchmark task source code are also released.}, number={9}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Hoang, Thanh Lam and Sbodio, Marco Luca and Martinez Galindo, Marcos and Zayats, Mykhaylo and Fernandez-Diaz, Raul and Valls, Victor and Picco, Gabriele and Berrospi, Cesar and Lopez, Vanessa}, year={2024}, month={Mar.}, pages={10544-10552} }