Login / Signup

Benchmarking Commercial Conformer Ensemble Generators.

Nils-Ole FriedrichChristina de Bruyn KopsFlorian FlachsenbergKai SommerMatthias RareyJohannes Kirchmair
Published in: Journal of chemical information and modeling (2017)
2017, 57, 529-539). For commercial algorithms, the median minimum root-mean-square deviations measured between protein-bound ligand conformations and ensembles of a maximum of 250 conformers are between 0.46 and 0.61 Å. Commercial conformer ensemble generators are characterized by their high robustness, with at least 99% of all input molecules successfully processed and few or even no substantial geometrical errors detectable in their output conformations. The RDKit distance geometry algorithm (with minimization enabled) appears to be a good free alternative since its performance is comparable to that of the midranked commercial algorithms. Based on a statistical analysis, we elaborate on which algorithms to use and how to parametrize them for best performance in different application scenarios.
Keyphrases
  • machine learning
  • deep learning
  • convolutional neural network
  • neural network
  • climate change
  • patient safety
  • small molecule
  • adverse drug