Scaffold hopping from natural products to synthetic mimetics by holistic molecular similarity

Abstract

Natural products offer unexplored molecular frameworks for the development of chemical leads and innovative drugs. However, the structural complexity of natural products compared with synthetic drug-like molecules often limits the scaffold hopping potential of naturalproduct-inspired molecular design. Here we introduce a holistic molecular representation incorporating pharmacophore and shape patterns, which facilitates scaffold hopping from natural products to isofunctional synthetic compounds. This computational approach captures simultaneously the partial charge, atom distributions and molecular shape. In a prospective application, we use four natural cannabinoids as queries in a chemical database search for novel synthetic modulators of human cannabinoid receptors. Of the synthetic compounds selected by the new method, 35% are experimentally confirmed as active. These cannabinoid receptor modulators are structurally less complex than their respective natural product templates. The results of this study validate this holistic molecular representation for hit and lead finding in drug discovery.

Discovery of Novel Molecular Frameworks of Farnesoid X Receptor Modulators by Ensemble Machine Learning

“…Each modeling method was selected as the result of a retrospective optimization procedure, in which we analyzed the performance of seven distinct molecular descriptions capturing 2D and 3D pharmacophore distributions (CATS2 and LIQUID, respectively), radial fragments (Morgan and FeatMorgan binary fingerprints), atom pairs (AtomPair fingerprints), topological and physicochemical properties (MOE2D descriptors), and molecular shape and partial charge distribution (WHALES) . The aggregation of predictors allows diverse molecular features responsible for the bioactivity to be taken into consideration, which thereby increases the overall predictive confidence.…”

“…MOE 2D descriptors were computed with MOE v. 2016.08 and default options. WHALES were calculated with open‐access Python software by using Gasteiger‐Marsili partial charges as a weighting scheme. LIQUID descriptors were computed with in‐house software with default settings.…”

New psychoactive substances: Pharmacology influencing UK practice, policy and the law

“…Synthetic cannabinoids are molecules that were designed to mimic and even improve the therapeutic effects of the plant product δ‐9 tetrahydrocannabinol (d9THC) . When the treatment potential of d9THC for nausea and vomiting and cachexia (weight loss) became accepted in the 1970s, a number of pharmaceutical companies decided to develop new molecules that, unlike d9THC, could be patented and then sold as medicines.…”

Design of Natural‐Product‐Inspired Multitarget Ligands by Machine Learning

“…(−)‐Galantamine ( 1 ) was used to computationally screen a library of 3 383 942 commercially available compounds. The partial charge distribution and three‐dimensional shape of compounds were captured using the weighted holistic atom localization and entity shape (WHALES) descriptors, which were specifically developed for scaffold‐hopping from natural products to synthetic compounds. WHALES capture the shape and partial charge distributions of a three‐dimensional (3D) molecular conformer in terms of atom‐centered Mahalanobis distances, weighted by atomic partial charges.…”

“…We employed the MMFF94 force field with 1000 iterations in a KNIME environment (‘RemoveStartingCoordinates’=True to ensure the reproducibility of the optimization procedure) and used the lowest‐energy conformer for each molecule for the subsequent WHALES descriptor calculation. WHALES values tolerate small conformational changes, thereby justifying the use of a single conformer for time‐efficient virtual screening.…”