Ligand-Target Interaction-Based Weighting of Substructures for Virtual Screening

Abstract

A methodology is introduced to assign energy-based scores to two-dimensional (2D) structural features based on three-dimensional (3D) ligand-target interaction information and utilize interaction-annotated features in virtual screening. Database molecules containing such fragments are assigned cumulative scores that serve as a measure of similarity to active reference compounds. The Interaction Annotated Structural Features (IASF) method is applied to mine five high-throughput screening (HTS) data sets and often identifies more hits than conventional fragment-based similarity searching or ligand-protein docking.

Molecular Interaction Fingerprints

“…The first attempt to restrict ligand fingerprints to protein interacting features was reported by Crisman et al [10] in the IASF (interaction annotated structural features) method that assigns precalculated interaction energy weights to ligand atoms in direct contact with the target. Starting from a protein-ligand X-ray structure, interaction energies are calculated on a per atom basis and summed over each ligand substructure encoded in a standard extended connectivity fingerprint [11].…”

Advances in 2D fingerprint similarity searching

“…This weight vector can then be incorporated into conventional Tc calculations, leading to a compound class-specific weighting scheme of the similarity calculations. Other recently introduced feature weighting schemes make indirect use of 3D ligand--target interaction information and are applicable to structural or molecular connectivity fingerprints [47][48][49][50]. Thus, these techniques bridge between 2D and 3D similarity searching on the basis of 2D fingerprint representations.…”

“…For each feature, a weight is derived on the basis of force field energy scores reflecting atomic contributions to interaction energies computed for crystallographic complexes. As a similarity value for a database compound relative to a ligand, the fingerprint features of a ligand contained in this pool are determined and their energy score weights are summed [47]. Despite the hypothetical nature of the interaction energies, the resulting feature sumbased database rankings have shown a significant enrichment of active compounds.…”

The multiple roles of computational chemistry in fragment-based drug design

“…Linking strategies were produced using CHARLIE, a part of the RACHEL de novo design suite in the Sybyl package, and QM calculations were performed to derive low energy conformations for the designed small molecules. Although not always applicable, fragment linking is a method that can be successful in a small number of cases [24]. Linker candidates were filtered based on the strain energy of the linked compound, the degree to which the linker distorted the original fragment binding mode and synthetic feasibility.…”