Screening Drug-Like Compounds by Docking to Homology Models:  A Systematic Study

Abstract

In the absence of an experimentally solved structure, a homology model of a protein target can be used instead for virtual screening of drug candidates by docking and scoring. This approach poses a number of questions regarding the choice of the template to use in constructing the model, the accuracy of the screening results, and the importance of allowing for protein flexibility. The present study addresses such questions with compound screening calculations for multiple homology models of five drug targets. A central result is that docking to homology models frequently yields enrichments of known ligands as good as that obtained by docking to a crystal structure of the actual target protein. Interestingly, however, standard measures of the similarity of the template used to build the homology model to the targeted protein show little correlation with the effectiveness of the screening calculations, and docking to the template itself often is as successful as docking to the corresponding homology model. Treating key side chains as mobile produces a modest improvement in the results. The reasons for these sometimes unexpected results, and their implications for future methodologic development, are discussed.

Homology modelling of the human adenosine A2B receptor based on X-ray structures of bovine rhodopsin, the β2-adrenergic receptor and the human adenosine A2A receptor

“…Indeed several examples in literature, suggest that homology modelling is a viable route to conduct, e.g. screening experiments by high throughput docking [19][20][21].…”

Virtual Screening and Biological Characterization of Novel Histone Arginine Methyltransferase PRMT1 Inhibitors

“…[27] It is surprising that homology models often provide significant enrichment of actives against a background of decoys. A variety of successful VS studies based [19,[28][29][30] In a retrospective study, Gilson and co-workers [31] analyzed five drug targets for which the crystal structure was available. The VS study showed that some homology models (based on a template with a sequence identity > 30 %) yielded enrichment equal to or even greater than that obtained with the crystal structure of the target.…”

Defensins and Bioinformatics: In Silico Approaches for Novel Therapeutic Antimicrobial Peptides

“…Some of these databases are made up of molecules that have the properties "drug-like" [98]. Moreover, such models may be used in simulations of "docking" together with the structure of active molecules identified in biological assays.…”