High-throughput screening (HTS) approaches have enabled rapid routine testing of millions of compounds towards the identification of novel “hit” molecules for therapeutic targets. However, many nuisance compounds, commonly labeled as PAINS, or Pan-Assay INterference compoundS, can interfere with the assay detection technology, leading to a false positive response and thus are of great interest and importance to the drug discovery community. In this study, we have generated, integrated, and curated large libraries of HTS data for the following most common interference assays: colloidal aggregation of beta-lactamase and cruzain, FireFly and Nano fluorescence imaging, thiol reactivity, and redox activity. We have developed several computational Quantitative Structure-Interference Relationship (QSIR) models of nuisance behavior-based on HTS data mainly designed for this study. These QSIR models, developed with several hundred thousand compounds tested under various experimental conditions, were developed using multiple machine learning algorithms, including both classic methods, such as Random Forest and Gradient Boosting Trees, as well as modern approaches, such as Multi-Descriptor Read Across and Deep Learning. All the algorithms exhibited similar behavior. We succeed in developing predictive models for all the assays, with correct classification rate (CCR) ranging between 60 and 80%. These models can be used in consensus to make an accurate and experimentally relevant prediction of potential nuisance behavior. In an effort to reduce wasted time and resource during pre-clinical drug discovery, we propose that these QSIR models be used to flag potential nuisance compounds instead of currently accepted and limited structural alert-based approaches. These models can be applied both prospectively during library design/annotation and retrospectively during HTS triage. We have developed an application tool for prediction of aggregation behavior named SCAM (Small, Colloidally-Aggregating Molecules) Detector, which we will present a live demonstration.

Accurate prediction of nuisance compounds based on large libraries of HTS data

Macromolecular Chemistry: The Second Century

[Author Benefits](https://scimeetings.acs.org/?utm_source=pubs_content_marketing&utm_medium=website&utm_campaign=0320_MCF_NPI_Launch_Spring_Homepage&ref=pubs_content_marketing)

[How to use SciMeetings](https://storage.googleapis.com/dl.morressier.com/acs/SciMeetings-ACSSpring2020-HowToGuide_M3020-Final.pdf)

[Frequently Asked Questions](https://storage.googleapis.com/dl.morressier.com/acs/SciMtng_ACSNMSp2020_FAQ_M3120_Final.pdf)

ACS Spring 2020 National Meeting & Expo

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.