Similarity-Quantized Relative Difference Learning for Improved Molecular Activity Prediction
Abstract
Accurate prediction of molecular activities is crucial for efficient drug discovery, yet remains challenging due to limited and noisy datasets. We introduce Similarity-Quantized Relative Learning (SQRL), a learning framework that reformulates molecular activity prediction as relative difference learning between structurally similar pairs of compounds. SQRL uses precomputed molecular similarities to enhance training of graph neural networks and other architectures, and significantly improves accuracy and generalization in low-data regimes common in drug discovery. We demonstrate its broad applicability and real-world potential through benchmarking on public datasets as well as proprietary industry data. Our findings demonstrate that leveraging similarity-aware relative differences provides an effective paradigm for molecular activity prediction.
Cite
Text
Zadorozhny et al. "Similarity-Quantized Relative Difference Learning for Improved Molecular Activity Prediction." NeurIPS 2024 Workshops: AIDrugX, 2024.Markdown
[Zadorozhny et al. "Similarity-Quantized Relative Difference Learning for Improved Molecular Activity Prediction." NeurIPS 2024 Workshops: AIDrugX, 2024.](https://mlanthology.org/neuripsw/2024/zadorozhny2024neuripsw-similarityquantized/)BibTeX
@inproceedings{zadorozhny2024neuripsw-similarityquantized,
title = {{Similarity-Quantized Relative Difference Learning for Improved Molecular Activity Prediction}},
author = {Zadorozhny, Karina and Chuang, Kangway V. and Sathappan, Bharath and Wallace, Ewan and Sresht, Vishnu and Grambow, Colin A},
booktitle = {NeurIPS 2024 Workshops: AIDrugX},
year = {2024},
url = {https://mlanthology.org/neuripsw/2024/zadorozhny2024neuripsw-similarityquantized/}
}