Drug Discovery under Covariate Shift with Domain-Informed Prior Distributions over Functions

Leo Klarner, Tim G. J. Rudner, Michael Reutlinger, Torsten Schindler, Garrett M Morris, Charlotte Deane, Yee Whye Teh
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:17176-17197, 2023.

Abstract

Accelerating the discovery of novel and more effective therapeutics is an important pharmaceutical problem in which deep learning is playing an increasingly significant role. However, real-world drug discovery tasks are often characterized by a scarcity of labeled data and significant covariate shift—a setting that poses a challenge to standard deep learning methods. In this paper, we present Q-SAVI, a probabilistic model able to address these challenges by encoding explicit prior knowledge of the data-generating process into a prior distribution over functions, presenting researchers with a transparent and probabilistically principled way to encode data-driven modeling preferences. Building on a novel, gold-standard bioactivity dataset that facilitates a meaningful comparison of models in an extrapolative regime, we explore different approaches to induce data shift and construct a challenging evaluation setup. We then demonstrate that using Q-SAVI to integrate contextualized prior knowledge of drug-like chemical space into the modeling process affords substantial gains in predictive accuracy and calibration, outperforming a broad range of state-of-the-art self-supervised pre-training and domain adaptation techniques.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-klarner23a, title = {Drug Discovery under Covariate Shift with Domain-Informed Prior Distributions over Functions}, author = {Klarner, Leo and Rudner, Tim G. J. and Reutlinger, Michael and Schindler, Torsten and Morris, Garrett M and Deane, Charlotte and Teh, Yee Whye}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {17176--17197}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/klarner23a/klarner23a.pdf}, url = {https://proceedings.mlr.press/v202/klarner23a.html}, abstract = {Accelerating the discovery of novel and more effective therapeutics is an important pharmaceutical problem in which deep learning is playing an increasingly significant role. However, real-world drug discovery tasks are often characterized by a scarcity of labeled data and significant covariate shift—a setting that poses a challenge to standard deep learning methods. In this paper, we present Q-SAVI, a probabilistic model able to address these challenges by encoding explicit prior knowledge of the data-generating process into a prior distribution over functions, presenting researchers with a transparent and probabilistically principled way to encode data-driven modeling preferences. Building on a novel, gold-standard bioactivity dataset that facilitates a meaningful comparison of models in an extrapolative regime, we explore different approaches to induce data shift and construct a challenging evaluation setup. We then demonstrate that using Q-SAVI to integrate contextualized prior knowledge of drug-like chemical space into the modeling process affords substantial gains in predictive accuracy and calibration, outperforming a broad range of state-of-the-art self-supervised pre-training and domain adaptation techniques.} }
Endnote
%0 Conference Paper %T Drug Discovery under Covariate Shift with Domain-Informed Prior Distributions over Functions %A Leo Klarner %A Tim G. J. Rudner %A Michael Reutlinger %A Torsten Schindler %A Garrett M Morris %A Charlotte Deane %A Yee Whye Teh %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-klarner23a %I PMLR %P 17176--17197 %U https://proceedings.mlr.press/v202/klarner23a.html %V 202 %X Accelerating the discovery of novel and more effective therapeutics is an important pharmaceutical problem in which deep learning is playing an increasingly significant role. However, real-world drug discovery tasks are often characterized by a scarcity of labeled data and significant covariate shift—a setting that poses a challenge to standard deep learning methods. In this paper, we present Q-SAVI, a probabilistic model able to address these challenges by encoding explicit prior knowledge of the data-generating process into a prior distribution over functions, presenting researchers with a transparent and probabilistically principled way to encode data-driven modeling preferences. Building on a novel, gold-standard bioactivity dataset that facilitates a meaningful comparison of models in an extrapolative regime, we explore different approaches to induce data shift and construct a challenging evaluation setup. We then demonstrate that using Q-SAVI to integrate contextualized prior knowledge of drug-like chemical space into the modeling process affords substantial gains in predictive accuracy and calibration, outperforming a broad range of state-of-the-art self-supervised pre-training and domain adaptation techniques.
APA
Klarner, L., Rudner, T.G.J., Reutlinger, M., Schindler, T., Morris, G.M., Deane, C. & Teh, Y.W.. (2023). Drug Discovery under Covariate Shift with Domain-Informed Prior Distributions over Functions. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:17176-17197 Available from https://proceedings.mlr.press/v202/klarner23a.html.

Related Material