Most drug-like molecules contain a number of rotatable bonds and prediction of bioactivities, docking etc. require an understanding of conformation. Whilst systematic methods can in theory explore all conformational space, however as the number of rotatable bonds increases a systematic search becomes prohibitive both in terms of the computational cost in generating conformations but also the time taken to process all the generated conformations (e.g. dock into protein). Thus there is great interest in rapid means to generate ensembles of representative conformations.
A recent paper DOI helps to address the problem by compiling a high quality dataset of structures generated using ligands from the Protein Data Bank.
The datasets were applied to benchmarking seven freely available conformer ensemble generators: Balloon (two different algorithms), the RDKit standard conformer ensemble generator, the Experimental-Torsion basic Knowledge Distance Geometry (ETKDG) algorithm, Confab, Frog2 and Multiconf-DOCK. Substantial differences in the performance of the individual algorithms were observed, with RDKit and ETKDG generally achieving a favorable balance of accuracy, ensemble size and runtime.
The dataset is freely available for download http://www.zbh.uni-hamburg.de/?id=628