On the relationship between molecular high-throughput screening hit rates and molecular descriptors

High-Throughput Screening (HTS) is widely used in the pharmaceutical industry to identify novel chemical starting points for drug discovery projects. The current study focuses on the relationship between molecular HTS hit rate and four common molecular descriptors: molecular lipophilicity (as described by ClogP), size (heavy atom count, HEV), degree of aromaticity content (Fsp3) and fraction of molecular framework (fMF). Molecular HTS hit rate is defined as the ratio between the number of times a compound was labeled as active and the number of times the compound was tested in a diverse set of HTS assays. Our results show that all four descriptors are correlated with molecular hit rate. Beta-binomial statistical models were built to model hit rates as a function of these descriptors and to quantify the influence of the descriptors on the molecular HTS hit rate. The study shows that among the four descriptors, ClogP has the largest influence on molecular hit rate. Fsp3 and HEV are also independently influencing the molecular hit rate. fMF has only minor influence beyond that it is correlated to the other descriptors. Higher degree polynomial terms of these descriptors were added into the beta-binomial statistic model to improve the model quality. The results show that models with high degree polynomial terms fit better to the experimental data than the non-polynomial models. In conclusion, for the first time the relative importance of the different molecular descriptors on the molecular HTS hit rate has been estimated. It was found that ClogP has the largest influence followed by Fsp3 and HEV.