Prediction of biological activity for high-throughput screening using binary kernel discrimination

High-throughput screening has made a significant impact on drug discovery, but there is an acknowledged need for quantitative methods to analyze scree ning results and predict the activity of further compounds. In this paper w e introduce one such method, binary kernel discrimination, and investigate its performance on two datasets; the first is a set of 1650 monoamine oxida se inhibitors, and the second a set of 101437 compounds from an in-house en zyme 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 per form robustly with varying quantities of training data and also in the pres ence of noisy data. We conclude by highlighting the importance of the judic ious use of,general pattern recognition techniques for compound selection.