FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313489214> ?p ?o ?g. }

• W4313489214 endingPage "458" @default.
• W4313489214 startingPage "442" @default.
• W4313489214 abstract "Although computational predictions of pharmacokinetics (PK) are desirable at the drug design stage, existing approaches are often limited by prediction accuracy and human interpretability. Using a discovery data set of mouse and rat PK studies at Roche (9,685 unique compounds), we performed a proof-of-concept study to predict key PK properties from chemical structure alone, including plasma clearance (CLp), volume of distribution at steady-state (Vss), and oral bioavailability (F). Ten machine learning (ML) models were evaluated, including Single-Task, Multitask, and transfer learning approaches (i.e., pretraining with in vitro data). In addition to prediction accuracy, we emphasized human interpretability of outcomes, especially the quantification of uncertainty, applicability domains, and explanations of predictions in terms of molecular features. Results show that intravenous (IV) PK properties (CLp and Vss) can be predicted with good precision (average absolute fold error, AAFE of 1.96–2.84 depending on data split) and low bias (average fold error, AFE of 0.98–1.36), with AutoGluon, Gaussian Process Regressor (GP), and ChemProp displaying the best performance. Driven by higher complexity of oral PK studies, predictions of F were more challenging, with the best AAFE values of 2.35–2.60 and higher overprediction bias (AFE of 1.45–1.62). Multi-Task approaches and pretraining of ChemProp neural networks with in vitro data showed similar precision to Single-Task models but helped reduce the bias and increase correlations between observations and predictions. A combination of GP-computed prediction variance, molecular clustering, and dimensionality-reduction provided valuable quantitative insights into prediction uncertainty and applicability domains. SHAPley Additive exPlanations (SHAPs) highlighted molecular features contributing to prediction outcomes of Vss, providing explanations that could aid drug design. Combined results show that computational predictions of PK are feasible at the drug design stage, with several ML technologies converging to successfully leverage historical PK data sets. Further studies are needed to unlock the full potential of this approach, especially with respect to data set sizes and quality, transfer learning between in vitro and in vivo data sets, model-independent quantification of uncertainty, and explainability of predictions." @default.
• W4313489214 created "2023-01-06" @default.
• W4313489214 creator A5001629429 @default.
• W4313489214 creator A5016570078 @default.
• W4313489214 creator A5025933102 @default.
• W4313489214 creator A5026902523 @default.
• W4313489214 creator A5034060219 @default.
• W4313489214 creator A5046488580 @default.
• W4313489214 creator A5059102184 @default.
• W4313489214 creator A5089231090 @default.
• W4313489214 date "2023-01-03" @default.
• W4313489214 modified "2023-10-11" @default.
• W4313489214 title "Computational Predictions of Nonclinical Pharmacokinetics at the Drug Design Stage" @default.
• W4313489214 cites W151377110 @default.
• W4313489214 cites W1932157650 @default.
• W4313489214 cites W1970707649 @default.
• W4313489214 cites W1974361103 @default.
• W4313489214 cites W1981737096 @default.
• W4313489214 cites W1988037271 @default.
• W4313489214 cites W1988909822 @default.
• W4313489214 cites W2004576565 @default.
• W4313489214 cites W2011997097 @default.
• W4313489214 cites W2012929832 @default.
• W4313489214 cites W2020859943 @default.
• W4313489214 cites W2046945713 @default.
• W4313489214 cites W2082804431 @default.
• W4313489214 cites W2086598386 @default.
• W4313489214 cites W2090234567 @default.
• W4313489214 cites W2126942199 @default.
• W4313489214 cites W2131326895 @default.
• W4313489214 cites W2150155198 @default.
• W4313489214 cites W2169237068 @default.
• W4313489214 cites W2564730406 @default.
• W4313489214 cites W2766856748 @default.
• W4313489214 cites W2803364122 @default.
• W4313489214 cites W2889326414 @default.
• W4313489214 cites W2897005416 @default.
• W4313489214 cites W2905674093 @default.
• W4313489214 cites W2911964244 @default.
• W4313489214 cites W2912518841 @default.
• W4313489214 cites W2937307539 @default.
• W4313489214 cites W2966357564 @default.
• W4313489214 cites W2968219050 @default.
• W4313489214 cites W2968734407 @default.
• W4313489214 cites W2988191536 @default.
• W4313489214 cites W3014339631 @default.
• W4313489214 cites W3023983310 @default.
• W4313489214 cites W3030834496 @default.
• W4313489214 cites W3041286284 @default.
• W4313489214 cites W3046713517 @default.
• W4313489214 cites W3083191400 @default.
• W4313489214 cites W3093687066 @default.
• W4313489214 cites W3095703932 @default.
• W4313489214 cites W3107587236 @default.
• W4313489214 cites W3110367110 @default.
• W4313489214 cites W3113124382 @default.
• W4313489214 cites W3119499511 @default.
• W4313489214 cites W3124422520 @default.
• W4313489214 cites W3125345344 @default.
• W4313489214 cites W3131648425 @default.
• W4313489214 cites W3157008596 @default.
• W4313489214 cites W3157530217 @default.
• W4313489214 cites W3189164715 @default.
• W4313489214 cites W3204865406 @default.
• W4313489214 cites W3211668464 @default.
• W4313489214 cites W3216996162 @default.
• W4313489214 cites W4210436046 @default.
• W4313489214 cites W4223546410 @default.
• W4313489214 cites W4225086330 @default.
• W4313489214 cites W4239082087 @default.
• W4313489214 cites W4282942502 @default.
• W4313489214 cites W4296844995 @default.
• W4313489214 doi "https://doi.org/10.1021/acs.jcim.2c01134" @default.
• W4313489214 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36595708" @default.
• W4313489214 hasPublicationYear "2023" @default.
• W4313489214 type Work @default.
• W4313489214 citedByCount "5" @default.
• W4313489214 countsByYear W43134892142023 @default.
• W4313489214 crossrefType "journal-article" @default.
• W4313489214 hasAuthorship W4313489214A5001629429 @default.
• W4313489214 hasAuthorship W4313489214A5016570078 @default.
• W4313489214 hasAuthorship W4313489214A5025933102 @default.
• W4313489214 hasAuthorship W4313489214A5026902523 @default.
• W4313489214 hasAuthorship W4313489214A5034060219 @default.
• W4313489214 hasAuthorship W4313489214A5046488580 @default.
• W4313489214 hasAuthorship W4313489214A5059102184 @default.
• W4313489214 hasAuthorship W4313489214A5089231090 @default.
• W4313489214 hasConcept C112705442 @default.
• W4313489214 hasConcept C119857082 @default.
• W4313489214 hasConcept C121955636 @default.
• W4313489214 hasConcept C144133560 @default.
• W4313489214 hasConcept C154945302 @default.
• W4313489214 hasConcept C167085575 @default.
• W4313489214 hasConcept C196083921 @default.
• W4313489214 hasConcept C2781067378 @default.
• W4313489214 hasConcept C2910466267 @default.
• W4313489214 hasConcept C41008148 @default.
• W4313489214 hasConcept C60644358 @default.

• next