FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385231489> ?p ?o ?g. }

• W4385231489 abstract "Abstract Explainable machine learning is increasingly used in drug discovery to help rationalize compound property predictions. Feature attribution techniques are popular choices to identify which molecular substructures are responsible for a predicted property change. However, established molecular feature attribution methods have so far displayed low performance for popular deep learning algorithms such as graph neural networks (GNNs), especially when compared with simpler modeling alternatives such as random forests coupled with atom masking. To mitigate this problem, a modification of the regression objective for GNNs is proposed to specifically account for common core structures between pairs of molecules. The presented approach shows higher accuracy on a recently-proposed explainability benchmark. This methodology has the potential to assist with model explainability in drug discovery pipelines, particularly in lead optimization efforts where specific chemical series are investigated." @default.
• W4385231489 created "2023-07-26" @default.
• W4385231489 creator A5034958835 @default.
• W4385231489 creator A5038390721 @default.
• W4385231489 creator A5054374232 @default.
• W4385231489 date "2023-07-25" @default.
• W4385231489 modified "2023-10-16" @default.
• W4385231489 title "Explaining compound activity predictions with a substructure-aware loss for graph neural networks" @default.
• W4385231489 cites W2019678805 @default.
• W4385231489 cites W2027376226 @default.
• W4385231489 cites W2033757486 @default.
• W4385231489 cites W2039609876 @default.
• W4385231489 cites W2047603685 @default.
• W4385231489 cites W2060531713 @default.
• W4385231489 cites W2096864392 @default.
• W4385231489 cites W2128428930 @default.
• W4385231489 cites W2134967712 @default.
• W4385231489 cites W2146844279 @default.
• W4385231489 cites W2168480393 @default.
• W4385231489 cites W2268755124 @default.
• W4385231489 cites W2295107390 @default.
• W4385231489 cites W2401258374 @default.
• W4385231489 cites W2517880718 @default.
• W4385231489 cites W2594183968 @default.
• W4385231489 cites W2606202972 @default.
• W4385231489 cites W2790808809 @default.
• W4385231489 cites W2916327454 @default.
• W4385231489 cites W2946617578 @default.
• W4385231489 cites W2953948143 @default.
• W4385231489 cites W2962858109 @default.
• W4385231489 cites W2966357564 @default.
• W4385231489 cites W2995345309 @default.
• W4385231489 cites W2998496395 @default.
• W4385231489 cites W3023042104 @default.
• W4385231489 cites W3093687066 @default.
• W4385231489 cites W3133900975 @default.
• W4385231489 cites W3173178613 @default.
• W4385231489 cites W3201988411 @default.
• W4385231489 cites W4200071670 @default.
• W4385231489 cites W4210436046 @default.
• W4385231489 cites W4220790374 @default.
• W4385231489 cites W4221114454 @default.
• W4385231489 cites W4221151574 @default.
• W4385231489 cites W4225252134 @default.
• W4385231489 cites W4225876201 @default.
• W4385231489 cites W4229060137 @default.
• W4385231489 cites W4283367795 @default.
• W4385231489 cites W4293563303 @default.
• W4385231489 cites W4300961340 @default.
• W4385231489 cites W4309258819 @default.
• W4385231489 cites W4310222865 @default.
• W4385231489 cites W4310603653 @default.
• W4385231489 cites W4313703134 @default.
• W4385231489 cites W4360992523 @default.
• W4385231489 doi "https://doi.org/10.1186/s13321-023-00733-9" @default.
• W4385231489 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37491407" @default.
• W4385231489 hasPublicationYear "2023" @default.
• W4385231489 type Work @default.
• W4385231489 citedByCount "1" @default.
• W4385231489 crossrefType "journal-article" @default.
• W4385231489 hasAuthorship W4385231489A5034958835 @default.
• W4385231489 hasAuthorship W4385231489A5038390721 @default.
• W4385231489 hasAuthorship W4385231489A5054374232 @default.
• W4385231489 hasBestOaLocation W43852314891 @default.
• W4385231489 hasConcept C111472728 @default.
• W4385231489 hasConcept C119857082 @default.
• W4385231489 hasConcept C124101348 @default.
• W4385231489 hasConcept C132525143 @default.
• W4385231489 hasConcept C13280743 @default.
• W4385231489 hasConcept C138885662 @default.
• W4385231489 hasConcept C154945302 @default.
• W4385231489 hasConcept C185798385 @default.
• W4385231489 hasConcept C189950617 @default.
• W4385231489 hasConcept C205649164 @default.
• W4385231489 hasConcept C2776401178 @default.
• W4385231489 hasConcept C41008148 @default.
• W4385231489 hasConcept C41895202 @default.
• W4385231489 hasConcept C50644808 @default.
• W4385231489 hasConcept C60644358 @default.
• W4385231489 hasConcept C68762167 @default.
• W4385231489 hasConcept C74187038 @default.
• W4385231489 hasConcept C80444323 @default.
• W4385231489 hasConcept C86803240 @default.
• W4385231489 hasConceptScore W4385231489C111472728 @default.
• W4385231489 hasConceptScore W4385231489C119857082 @default.
• W4385231489 hasConceptScore W4385231489C124101348 @default.
• W4385231489 hasConceptScore W4385231489C132525143 @default.
• W4385231489 hasConceptScore W4385231489C13280743 @default.
• W4385231489 hasConceptScore W4385231489C138885662 @default.
• W4385231489 hasConceptScore W4385231489C154945302 @default.
• W4385231489 hasConceptScore W4385231489C185798385 @default.
• W4385231489 hasConceptScore W4385231489C189950617 @default.
• W4385231489 hasConceptScore W4385231489C205649164 @default.
• W4385231489 hasConceptScore W4385231489C2776401178 @default.
• W4385231489 hasConceptScore W4385231489C41008148 @default.
• W4385231489 hasConceptScore W4385231489C41895202 @default.
• W4385231489 hasConceptScore W4385231489C50644808 @default.
• W4385231489 hasConceptScore W4385231489C60644358 @default.
• W4385231489 hasConceptScore W4385231489C68762167 @default.
• W4385231489 hasConceptScore W4385231489C74187038 @default.

• next