BackgroundHalogen bonding has recently come to play as a target for lead optimization in rational drug design. However, most docking program don’t account for halogen bonding in their scoring functions and are not able to utilize this new approach. In this study a new and improved halogen bonding scoring function (XBSF) is presented along with its implementation in the AutoDock Vina molecular docking software. This new improved program is termed as AutoDock VinaXB, where XB stands for the halogen bonding parameters that were added.ResultsXBSF scoring function is derived based on the X···A distance and C–X···A angle of interacting atoms. The distance term was further corrected to account for the polar flattening effect of halogens. A total of 106 protein-halogenated ligand complexes were tested and compared in terms of binding affinity and docking poses using Vina and VinaXB. VinaXB performed superior to Vina in the majority of instances. VinaXB was closer to native pose both above and below 2 Å deviation categories almost twice as frequently as Vina.ConclusionsImplementation of XBSF into AutoDock Vina has been shown to improve the accuracy of the docking result with regards to halogenated ligands. AutoDock VinaXB addresses the issues of halogen bonds that were previously being scored unfavorably due to repulsion factors, thus effectively lowering the output RMSD values.Electronic supplementary materialThe online version of this article (doi:10.1186/s13321-016-0139-1) contains supplementary material, which is available to authorized users.

Sulfur bonding interactions between organosulfur compounds and proteins were examined using crystal structures deposited to-date in the PDB. The data was analyzed as a function of sulfur-σ-hole-bonding (i.e., sulfur bonds) to main chain Lewis bases, viz. oxygen and nitrogen atoms of the backbone amide linkages. The analyses also included an examination of sulfur bonding to side chain Lewis bases (O, N, and S) and to the "non-classical" Lewis bases present in electron-rich aromatic amino acids as-well-as to donor-acceptor bond angle distributions. The interactions analyzed included those restricted to the sum of van der Waals radii of the respective atoms or to a distance of 4 Å. The surveyed data revealed that sulfur bonding tendencies (C-S-C bond angles) were impacted not only by steric effects but perhaps also by enthalpic features present in both the donor and acceptor participants. This knowledge is not only of fundamental interest but is also important in terms of materials and drug-design involving moieties incorporating the sulfur atom. Additionally, a new empirical scoring function was developed to address the anisotropy of sulfur in protein-ligand interactions. This newly developed scoring function is incorporated into AutoDock Vina molecular docking program and is valuable for modeling and drug design.