FLEXS - A METHOD FOR FAST FLEXIBLE LIGAND SUPERPOSITION

If no structural information about a particular target protein is avai lable, methods of rational drug design try to superimpose putative lig ands with a given reference, e.g., an endogenous ligand. The goal of s uch structural alignments is, on the one hand, to approximate the bind ing geometry and, on the other hand, to provide a relative ranking of the ligands with respect to their similarity. An accurate superpositio n is the prerequisite of subsequent exploitation of ligand data by eit her 3D QSAR analyses, pharmacophore hypotheses, or receptor modeling. We present the automatic method FLEXS for structurally superimposing p airs of ligands, approximating their putative binding site geometry. O ne of the ligands is treated as flexible, while the other one, used as a reference, is kept rigid. FLEXS is an incremental construction proc edure. The molecules to be superimposed are partitioned into fragments . Starting with placements of a selected anchor fragment, computed by two alternative approaches, the remaining fragments are added iterativ ely. At each step, flexibility is considered by allowing the respectiv e added fragment to adopt a discrete set of conformations. The mean co mputing time per test case is about 1:30 min on a common-day workstati on. FLEXS is fast enough to be used as a tool for virtual ligand scree ning. A database of typical drug molecules has been screened for poten tial fibrinogen receptor antagonists. FLEXS is capable of retrieving a ll ligands assigned to platelet aggregation properties among the first 20 hits. Furthermore, the program suggests additional interesting can didates, likely to be active at the same receptor. FLEXS proves to be superior to commonly used retrieval techniques based on 2D fingerprint similarities. The accuracy of computed superpositions determines the relevance of subsequently performed ligand analyses. In order to valid ate the quality of FLEXS alignments, we attempted to reproduce a set o f 284 mutual superpositions derived from experimental data on 76 prote in-ligand complexes of 14 proteins. The ligands considered cover the w hole range of drug-size molecules from 18 to 158 atoms (PDB codes: 3pt b, 2er7). The performance of the algorithm critically depends on the s izes of the molecules to be superimposed. The limitations are clearly demonstrated with large peptidic inhibitors in the HIV and the endothi apepsin data set. Problems also occur in the presence of multiple bind ing modes (e.g., elastase and human rhinovirus). The most convincing r esults are achieved with small- and medium-sized molecules (as, e.g., the ligands of trypsin, thrombin, and dihydrofolate reductase). In mor e than half of the entire test set, we achieve rms deviations between computed and observed alignment of below 1.5 Angstrom. This underlines the reliability of FLEXS-generated alignments.