Motivation-Accurately predicting the binding affinities of large sets of diverse protein-ligand complexes is an extremely challenging task. The scoring functions that attempt such computational prediction are essential for analysing the outputs of Molecular Docking, which is in turn an important technique for drug discovery, chemical biology and structural biology. Each scoring function assumes a predetermined theory-inspired functional form for the relationship between the variables that characterise the complex, which also include parameters fitted to experimental or simulation data, and its predicted binding affinity. The inherent problem of this rigid approach is that it leads to poor predictivity for those complexes that do not conform to the modelling assumptions. Moreover, resampling strategies, such as cross-validation or bootstrapping, are still not systematically used to guard against the overfitting of calibration data in parameter estimation for scoring functions.Results-We propose a novel scoring function (RF-Score) that circumvents the need for problematic modelling assumptions via non-parametric machine learning. In particular, Random Forest was used to implicitly capture binding effects that are hard to model explicitly. RF-Score is compared with the state of the art on the demanding PDBbind benchmark. Results show that RFScore is a very competitive scoring function. Importantly, RF-Score's performance was shown to improve dramatically with training set size and hence the future availability of more high quality structural and interaction data is expected to lead to improved versions of RF-Score.

Machine learning algorithms are generally developed in computer science or adjacent disciplines and find their way into chemical modeling by a process of diffusion. Though particular machine learning methods are popular in chemoinformatics and quantitative structure–activity relationships (QSAR), many others exist in the technical literature. This discussion is methods-based and focused on some algorithms that chemoinformatics researchers frequently use. It makes no claim to be exhaustive. We concentrate on methods for supervised learning, predicting the unknown property values of a test set of instances, usually molecules, based on the known values for a training set. Particularly relevant approaches include Artificial Neural Networks, Random Forest, Support Vector Machine, k-Nearest Neighbors and naïve Bayes classifiers.

Over the past decade, pharmaceutical companies have seen a decline in the number of drug candidates successfully passing through clinical trials, though billions are still spent on drug development. Poor aqueous solubility leads to low bio-availability, reducing pharmaceutical effectiveness. The human cost of inefficient drug candidate testing is of great medical concern, with fewer drugs making it to the production line, slowing the development of new treatments. In biochemistry and biophysics, water mediated reactions and interactions within active sites and protein pockets are an active area of research, in which methods for modelling solvated systems are continually pushed to their limits. Here, we discuss a multitude of methods aimed towards solvent modelling and solubility prediction, aiming to inform the reader of the options available, and outlining the various advantages and disadvantages of each approach.