Predicting Endocrine Disruption and Blood-Brain Barrier Permeability using the EVANS QSPR Methodology
This work presents the development and application of a Quantitative Structure Property Relationship (QSPR) methodology entitled ‘EigenValue ANalysiS (EVANS)’. The EVANS methodology combines information extracted from 3D molecular structures with 2D physicochemical properties to generate eigenvalues, which are then used as descriptor variables in QSPR analyses. Having previously benchmarked the methodology on pharmacodynamic and pharmacokinetic data, we present the EVANS formalism in brief and deploy it on high-quality datasets for prediction of two pharmacological endpoints: (i) estrogen receptor (ER) mediated endocrine disruption, and (ii) BBB permeability. The datasets were processed through the methodology and the generated eigenvalue descriptors were used in conjunction with machine learning algorithms to build predictive QSPR models. Our results demonstrate that the EVANS methodology produces robust models that meet the standards of a new QSPR formalism and can be a powerful tool to design safe and effective drugs.