Identification of Compounds That Interfere with High‐Throughput Screening Assay Technologies. (19th September 2019)

• Record Type: Journal Article
• Authors: David, Laurianne; Walsh, Jarrod; Sturm, Noé; Feierberg, Isabella; Nissink, J. Willem M.; Chen, Hongming; Bajorath, Jürgen; Engkvist, Ola
• Is Part Of: ChemMedChem. Volume 14:Number 20(2019)
• Page Start: 1795
• Page End: 1802
• Publication Date: 2019-09-19
• Subjects: assay interference -- computational chemistry -- frequent hitters -- high-throughput screening -- machine learning; Pharmaceutical chemistry -- Periodicals; 615.19005
• DOI: 10.1002/cmdc.201900395
• Languages: English
• ISSNs: 1860-7179

Abstract

Abstract: A significant challenge in high‐throughput screening (HTS) campaigns is the identification of assay technology interference compounds. AC ompoundI nterfering with anA ssayT echnology (CIAT) gives false readouts in many assays. CIATs are often considered viable hits and investigated in follow‐up studies, thus impeding research and wasting resources. In this study, we developed a machine‐learning (ML) model to predict CIATs for three assay technologies. The model was trained on known CIATs and non‐CIATs (NCIATs) identified in artefact assays and described by their 2D structural descriptors. Usual methods identifying CIATs are based on statistical analysis of historical primary screening data and do not consider experimental assays identifying CIATs. Our results show successful prediction of CIATs for existing and novel compounds and provide a complementary and wider set of predicted CIATs compared to BSF, a published structure‐independent model, and to the PAINS substructural filters. Our analysis is an example of how well‐curated datasets can provide powerful predictive models despite their relatively small size. Abstract : Prediction of compounds that interfere with HTS technology through a machine‐learning model: In this work, compounds that interfere with a high‐throughput screening technology are identified in counter‐screen assays and their chemical structures are used to train a random‐forest model to predict the behavior of new compounds. The model performs well, with respective ROC AUC values of 0.70, 0.62, and 0.57 in AlphaScreen, FRET, and TR‐FRET technologies and outperforms another published statistical method.