Machine learning prediction of 3CLpro SARS-CoV-2 docking scores.

Molecular docking results of two training sets containing 866 and 8,696 compounds were used to train three different machine learning (ML) approaches. Neural network approaches according to Keras and TensorFlow libraries and the gradient boosted decision trees approach of XGBoost were used with DScribe's Smooth Overlap of Atomic Positions molecular descriptors. In addition, neural networks using the SchNetPack library and descriptors were used. The ML performance was tested on three different sets, including compounds for future organic synthesis. The final evaluation of the ML predicted docking scores was based on the ZINC in vivo set, from which 1,200 compounds were randomly selected with respect to their size. The results obtained showed a consistent ML prediction capability of docking scores, and even though compounds with more than 60 atoms were found slightly overestimated they remain valid for a subsequent evaluation of their drug repurposing suitability.

Mass spectrometrical and quantum-chemical study of pentafluorophenylhydrazones.

Twenty-one pentafluorphenylhydrazones have been analyzed by means of tandem mass spectrometry (ESI MS/MS) conditions to compare their fragmentations with those ones obtained from quantum-chemical calculations of the hydrazone moiety depending on the substitution from the aldehyde site. The hydrazone N-N bond is disrupted under such conditions, and these results are in accordance with the facts that an electron-rich particle, such as an anion and or radical in a solution, can cause this disruption and simultaneous defluorination in para-position of the hydrazone part of the molecule.