Task-Similarity is a Crucial Factor for Few-Shot Meta-Learning of Structure-Activity Relationships.

Autor: Kötter A; R&D, Integrated Drug Discovery, Sanofi-Aventis Deutschland GmbH, Industriepark Höchst, 65926, Frankfurt am Main, Germany., Allenspach S; Department of Chemistry and Applied Biosciences, ETH Zurich, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland., Grebner C; R&D, Integrated Drug Discovery, Sanofi-Aventis Deutschland GmbH, Industriepark Höchst, 65926, Frankfurt am Main, Germany., Matter H; R&D, Integrated Drug Discovery, Sanofi-Aventis Deutschland GmbH, Industriepark Höchst, 65926, Frankfurt am Main, Germany., Hiss JA; Department of Chemistry and Applied Biosciences, ETH Zurich, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland., Schneider G; Department of Chemistry and Applied Biosciences, ETH Zurich, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland., Hessler G; R&D, Integrated Drug Discovery, Sanofi-Aventis Deutschland GmbH, Industriepark Höchst, 65926, Frankfurt am Main, Germany.
Jazyk: angličtina
Zdroj: Chembiochem : a European journal of chemical biology [Chembiochem] 2024 Oct 01; Vol. 25 (19), pp. e202400095. Date of Electronic Publication: 2024 Jun 18.
DOI: 10.1002/cbic.202400095
Abstrakt: Machine learning models support computer-aided molecular design and compound optimization. However, the initial phases of drug discovery often face a scarcity of training data for these models. Meta-learning has emerged as a potentially promising strategy, harnessing the wealth of structure-activity data available for known targets to facilitate efficient few-shot model training for the specific target of interest. In this study, we assessed the effectiveness of two different meta-learning methods, namely model-agnostic meta-learning (MAML) and adaptive deep kernel fitting (ADKF), specifically in the regression setting. We investigated how factors such as dataset size and the similarity of training tasks impact predictability. The results indicate that ADKF significantly outperformed both MAML and a single-task baseline model on the inhibition data. However, the performance of ADKF varied across different test tasks. Our findings suggest that considerable enhancements in performance can be anticipated primarily when the task of interest is similar to the tasks incorporated in the meta-learning process.
(© 2024 Wiley-VCH GmbH.)
Databáze: MEDLINE
Externí odkaz: