Show simple item record

dc.contributor.authorIrwin, Benedict W. J.
dc.contributor.authorWhitehead, Thomas M.
dc.contributor.authorRowland, Scott
dc.contributor.authorMahmoud, Samar Y.
dc.contributor.authorConduit, Gareth J.
dc.contributor.authorSegall, Matthew D.
dc.date.accessioned2021-07-07T15:47:09Z
dc.date.available2021-07-07T15:47:09Z
dc.date.issued2021-07-07
dc.date.submitted2021-01-13
dc.identifier.issn2689-5595
dc.identifier.otherail231
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/324934
dc.description.abstractAbstract: More accurate predictions of the biological properties of chemical compounds would guide the selection and design of new compounds in drug discovery and help to address the enormous cost and low success‐rate of pharmaceutical R&D. However, this domain presents a significant challenge for AI methods due to the sparsity of compound data and the noise inherent in results from biological experiments. In this paper, we demonstrate how data imputation using deep learning provides substantial improvements over quantitative structure‐activity relationship (QSAR) machine learning models that are widely applied in drug discovery. We present the largest‐to‐date successful application of deep‐learning imputation to datasets which are comparable in size to the corporate data repository of a pharmaceutical company (678 994 compounds by 1166 endpoints). We demonstrate this improvement for three areas of practical application linked to distinct use cases; (a) target activity data compiled from a range of drug discovery projects, (b) a high value and heterogeneous dataset covering complex absorption, distribution, metabolism, and elimination properties, and (c) high throughput screening data, testing the algorithm's limits on early stage noisy and very sparse data. Achieving median coefficients of determination, R2, of 0.69, 0.36, and 0.43, respectively, across these applications, the deep learning imputation method offers an unambiguous improvement over random forest QSAR methods, which achieve median R2 values of 0.28, 0.19, and 0.23, respectively. We also demonstrate that robust estimates of the uncertainties in the predicted values correlate strongly with the accuracies in prediction, enabling greater confidence in decision‐making based on the imputed values.
dc.languageen
dc.publisherBlackwell Publishing Ltd
dc.subjectLETTER
dc.subjectLETTERS
dc.subjectAI in drug discovery
dc.subjectdeep learning
dc.subjectdrug discovery
dc.subjectimputation
dc.titleDeep imputation on large‐scale drug discovery data
dc.typeOther
dc.date.updated2021-07-07T15:47:08Z
prism.publicationNameApplied AI Letters
dc.identifier.doi10.17863/CAM.72387
dcterms.dateAccepted2021-05-18
rioxxterms.versionofrecord10.1002/ail2.31
rioxxterms.versionAO
rioxxterms.versionVoR
rioxxterms.licenseref.urihttp://creativecommons.org/licenses/by/4.0/
dc.contributor.orcidIrwin, Benedict W. J. [0000-0001-5102-7439]


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record