Designing Focused Chemical Libraries Enriched in Protein-Protein Interaction Inhibitors using Machine-Learning Methods

Protein-protein interactions (PPIs) may represent one of the next major classes of therapeutic targets. So far, only a minute fraction of the estimated 650,000 PPIs that comprise the human interactome are known with a tiny number of complexes being drugged. Such intricate biological systems cannot be cost-efficiently tackled using conventional high-throughput screening methods. Rather, time has come for designing new strategies that will maximize the chance for hit identification through a rationalization of the PPI inhibitor chemical space and the design of PPI-focused compound libraries (global or target-specific). Here, we train machine-learning-based models, mainly decision trees, using a dataset of known PPI inhibitors and of regular drugs in order to determine a global physico-chemical profile for putative PPI inhibitors. This statistical analysis unravels two important molecular descriptors for PPI inhibitors characterizing specific molecular shapes and the presence of a privileged number of aromatic bonds. The best model has been transposed into a computer program, PPI-HitProfiler, that can output from any drug-like compound collection a focused chemical library enriched in putative PPI inhibitors. Our PPI inhibitor profiler is challenged on the experimental screening results of 11 different PPIs among which the p53/MDM2 interaction screened within our own CDithem platform, that in addition to the validation of our concept led to the identification of 4 novel p53/MDM2 inhibitors. Collectively, our tool shows a robust behavior on the 11 experimental datasets by correctly profiling 70% of the experimentally identified hits while removing 52% of the inactive compounds from the initial compound collections. We strongly believe that this new tool can be used as a global PPI inhibitor profiler prior to screening assays to reduce the size of the compound collections to be experimentally screened while keeping most of the true PPI inhibitors. PPI-HitProfiler is freely available on request from our CDithem platform website, www.CDithem.com.Author Summary: Protein-protein interactions (PPIs) are essential to life and various diseases states are associated with aberrant PPIs. Therefore significant efforts are dedicated to this new class of therapeutic targets. Even though it might not be possible to modulate the estimated 650,000 PPIs that regulate human life with drug-like compounds, a sizeable number of PPI should be druggable. Only 10-15% of the human genome is thought to be druggable with around 1000-3000 druggable protein targets. A hypothetical similar ratio for PPIs would bring the number of druggable PPIs to about 65,000, although no data can yet support such a hypothesis. PPI have been historically intricate to tackle with standard experimental and virtual screening techniques, possibly because of the shift in the chemical space between today's chemical libraries and PPI physico-chemical requirements. Therefore, one possible avenue to circumvent this conundrum is to design focused libraries enriched in putative PPI inhibitors. Here, we show how chemoinformatics can assist library design by learning physico-chemical rules from a data set of known PPI inhibitors and their comparison with regular drugs. Our study shows the importance of specific molecular shapes and a privileged number of aromatic bonds.