arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Computer Science · Computational Theory and Mathematics

Application of Validated QSAR Models of D<sub>1</sub> Dopaminergic Antagonists for\nDatabase Mining

Scott Oloff (2517334), Richard B. Mailman (170558), Alexander Tropsha (1308912)

· Figshare

See on the map

Abstract

Rigorously validated quantitative structure−activity relationship (QSAR) models have been\ndeveloped for 48 antagonists of the dopamine D<sub>1</sub> receptor and applied to mining chemical\ndatasets to discover novel potential antagonists. Several QSAR methods have been employed,\nincluding comparative molecular field analysis (CoMFA), simulated annealing−partial least\nsquares (SA-PLS), <i>k</i>-nearest neighbor (kNN), and support vector machines (SVM). With the\nexception of CoMFA, these approaches employed 2D topological descriptors generated with\nthe MolConnZ software package (EduSoft, LLC. MolconnZ, version 4.05; http://www.eslc.vabiotech.com/ [4.05], 2003). The original dataset was split into training and test sets to allow\nfor external validation of each training set model. The resulting models were characterized by\ncross-validated <i>R</i><sup>2</sup> (<i>q</i><sup>2</sup>) for the training set and predictive <i>R</i><sup>2</sup> values for the test set of (<i>q</i><sup>2</sup>/<i>R</i><sup>2</sup>)\n0.51/0.47 for CoMFA, 0.7/0.76 for kNN, <i>R</i><sup>2</sup> for the training and test sets of 0.74/0.71 for SVM,\nand training set fitness and test set <i>R</i><sup>2</sup> values of 0.68/0.63 for SA-PLS. Validated QSAR models\nwith <i>R</i><sup>2</sup> > 0.7, (i.e., kNN and SVM) were used to mine three publicly available chemical\ndatabases: the National Cancer Institute (NCI) database of ca. 250 000 compounds, the\nMaybridge Database of ca. 56 000 compounds, and the ChemDiv Database of ca. 450 000\ncompounds. These searches resulted in only 54 <i>consensus</i> hits (i.e., predicted active by all\nmodels); five of them were previously characterized as dopamine D<sub>1</sub> ligands, but were not\npresent in the original dataset. A small fraction of the purported D<sub>1</sub> ligands did not contain a\ncatechol ring found in all known dopamine full agonist ligands, suggesting that they may be\nnovel structural antagonist leads. This study illustrates that the combined application of\npredictive QSAR modeling and database mining may provide an important avenue for rational\ncomputer-aided drug discovery.

More like this