Please use this identifier to cite or link to this item:

TitleEvaluating molecular representations in machine learning models for drug response prediction and interpretability
Author(s)Baptista, Delora
Correia, João
Pereira, Bruno
Rocha, Miguel
deep learning
drug sensitivity
learned representations
molecular fingerprints
Issue date26-Aug-2022
PublisherDe Gruyter
JournalJournal of Integrative Bioinformatics
CitationBaptista, Delora, Correia, João, Pereira, Bruno and Rocha, Miguel. "Evaluating molecular representations in machine learning models for drug response prediction and interpretability" Journal of Integrative Bioinformatics, vol. 19, no. 3, 2022, pp. 20220006.
Abstract(s)Machine learning (ML) is increasingly being used to guide drug discovery processes. When applying ML approaches to chemical datasets, molecular descriptors and fingerprints are typically used to represent compounds as numerical vectors. However, in recent years, end-to-end deep learning (DL) methods that can learn feature representations directly from line notations or molecular graphs have been proposed as alternatives to using precomputed features. This study set out to investigate which compound representation methods are the most suitable for drug sensitivity prediction in cancer cell lines. Twelve different representations were benchmarked on 5 compound screening datasets, using DeepMol, a new chemoinformatics package developed by our research group, to perform these analyses. The results of this study show that the predictive performance of end-to-end DL models is comparable to, and at times surpasses, that of models trained on molecular fingerprints, even when less training data is available. This study also found that combining several compound representation methods into an ensemble can improve performance. Finally, we show that a post hoc feature attribution method can boost the explainability of the DL models.
Publisher version
AccessOpen access
Appears in Collections:CEB - Publicações em Revistas/Séries Internacionais / Publications in International Journals/Series

Files in This Item:
File Description SizeFormat 
document_55770_1.pdf2,14 MBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons

Partilhe no FacebookPartilhe no TwitterPartilhe no DeliciousPartilhe no LinkedInPartilhe no DiggAdicionar ao Google BookmarksPartilhe no MySpacePartilhe no Orkut
Exporte no formato BibTex mendeley Exporte no formato Endnote Adicione ao seu ORCID