Matched pairs demonstrate robustness against inter-assay variability

• Published in: Journal of cheminformatics Vol. 17; no. 1; p. 8
• Main Authors: Nelen, Jochem, Pérez-Sánchez, Horacio, De Winter, Hans, Van Rompaey, Dries
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 20.01.2025; BioMed Central Ltd; Springer Nature B.V; BMC
• Subjects: Assay noise; Assaying; Brief Report; ChEMBL; Chemistry; Chemistry and Materials Science; Computational Biology/Bioinformatics; Computer Applications in Chemistry; Data curation; Datasets; Documentation and Information in Chemistry; Error reduction; Impact analysis; Information management; Machine learning; Matched structural pairs; Metadata; Noise reduction; Quality assessment; Quality control; Theoretical and Computational Chemistry; Data curation; ChEMBL; Matched structural pairs; Assay noise; Machine learning

Machine learning models for chemistry require large datasets, often compiled by combining data from multiple assays. However, combining data without careful curation can introduce significant noise. While absolute values from different assays are rarely comparable, trends or differences between comp...