Using machine learning to improve ensemble docking for drug discovery.

Ensemble docking has provided an inexpensive method to account for receptor flexibility in molecular docking for virtual screening. Unfortunately, as there is no rigorous theory to connect the docking scores from multiple structures to measured activity, researchers have not yet come up with effective ways to use these scores to classify compounds into actives and inactives. This shortcoming has led to the decrease, rather than an increase in the performance of classifying compounds when more structures are added to the ensemble. Previously, we suggested machine learning, implemented in the form of a naïve Bayesian model could alleviate this problem. However, the naïve Bayesian model assumed that the probabilities of observing the docking scores to different structures to be independent. This approximation might prevent it from achieving even higher performance. In the work presented in this paper, we have relaxed this approximation when using several other machine learning methods-k nearest neighbor, logistic regression, support vector machine, and random forest-to improve ensemble docking. We found significant improvement.

Spectroscopic investigation and direct comparison of the reactivities of iron pyridyl oxidation catalysts.

The growing interest in green chemistry has fueled attention to the development and characterization of effective iron complex oxidation catalysts. A number of iron complexes are known to catalyze the oxidation of organic substrates utilizing peroxides as the oxidant. Their development is complicated by a lack of direct comparison of the reactivities of the iron complexes. To begin to correlate reactivity with structural elements, we compare the reactivities of a series of iron pyridyl complexes toward a single dye substrate, malachite green (MG), for which colorless oxidation products are established. Complexes with tetradentate, nitrogen-based ligands with cis open coordination sites were found to be the most reactive. While some complexes reflect sensitivity to different peroxides, others are similarly reactive with either H2O2 or tBuOOH, which suggests some mechanistic distinctions. [Fe(S,S-PDP)(CH3CN)2](SbF6)2 and [Fe(OTf)2(tpa)] transition under the oxidative reaction conditions to a single intermediate at a rate that exceeds dye degradation (PDP=bis(pyridin-2-ylmethyl) bipyrrolidine; tpa=tris(2-pyridylmethyl)amine). For the less reactive [Fe(OTf)2(dpa)] (dpa=dipicolylamine), this reaction occurs on a timescale similar to that of MG oxidation. Thus, the spectroscopic method presented herein provides information about the efficiency and mechanism of iron catalyzed oxidation reactions as well as about potential oxidative catalyst decomposition and chemical changes of the catalyst before or during the oxidation reaction.