Using a simple model of ligand-receptor interactions, the interactions between ligands and receptors of varying complexities are studied and the probabilities of binding calculated. It is observed that as the systems become more complex the chance of observing a useful interaction for a randomly chosen ligand falls dramatically. The implications of this for the design of combinatorial libraries is explored. A large set of drug leads and optimized compounds is profiled using several different properties relevant to molecular recognition. The changes observed for these properties during the drug optimization phase support the hypothesis that less complex molecules are more common starting points for the discovery of drugs. An extreme example of the use of simple molecules for directed screening against thrombin is provided.

More than 500 compounds chosen to represent kinase inhibitor space have been screened against a panel of over 200 protein kinases. Significant results include the identification of hits against new kinases including PIM1 and MPSK1, and the expansion of the inhibition profiles of several literature compounds. A detailed analysis of the data through the use of affinity fingerprints has produced findings with implications for biological target selection, the choice of tool compounds for target validation, and lead discovery and optimization. In a detailed examination of the tyrosine kinases, interesting relationships have been found between targets and compounds. Taken together, these results show how broad cross-profiling can provide important insights to assist kinase drug discovery.

High-throughput screening has made a significant impact on drug discovery, but there is an acknowledged need for quantitative methods to analyze screening results and predict the activity of further compounds. In this paper we introduce one such method, binary kernel discrimination, and investigate its performance on two datasets; the first is a set of 1650 monoamine oxidase inhibitors, and the second a set of 101 437 compounds from an in-house enzyme assay. We compare the performance of binary kernel discrimination with a simple procedure which we call "merged similarity search", and also with a feedforward neural network. Binary kernel discrimination is shown to perform robustly with varying quantities of training data and also in the presence of noisy data. We conclude by highlighting the importance of the judicious use of general pattern recognition techniques for compound selection.