Orthologue chemical space and its influence on target prediction

• DOI: 10.1093/bioinformatics/btx525
• Publication Date: 2017-08-25T11:09:17Z
• Authors: Lewis H Mervin; Krishna C Bulusu; Leen Kalash; Avid M Afzal; Fredrik Svensson; Mike A Firth; Ian Barrett; Ola Engkvist; Andreas Bender
• Keywords: Chemical space; chEMBL; PubChem; Similarity (geometry)

In silico approaches often fail to utilize bioactivity data available for orthologous targets due insufficient evidence highlighting the benefit such an approach. Deeper investigation into orthologue chemical space and its influence toward expanding compound target coverage is necessary improve confidence in this practice.Here we present analysis of ChEMBL PubChem impact on prediction. We highlight number conflicting bioactivities between human orthologues low annotations are overall compatible. Chemical shows chemically dissimilar with high intra-group similarity, suggesting they could effectively extend modelled. Based these observations, show inclusion terms novel coverage. also benchmarked predictive models using a time-series split from Chemistry Connect HTS at AstraZeneca, showing that statistically improved performance.Orthologue-based prediction training set www.github.com/lhm30/[e-mail] Bioinformatics online.