Statistical mechanics with quantum potentials: Application to protein-ligand binding affinities

During the drug development processlead optimization requires intensive chemical synthesis and testing effortsThe process can be highly iterative in nature with multiple rounds of synthesis requiredbecause changes made to improvefor examplepharmacokinetic factors such as solubility can also decrease potencyrequiring further changes to recover potencyand so onConsistently accurate computational predictions of protein ligand binding affinities would significantly reduce this expensive and time consuming burdenby providing medicinal chemists the ability to more aggressively prioritize ligands for synthesis and testing based on computational resultsHowevercurrentlyachievement of consistent accuracy in protein ligand binding affinity prediction is an unmet goal in the field of computational chemistryConventional docking and scoring methods have been shown to provide enrichment of active vsinactive ligands in chemical librariesbut still are very limited in their ability to rank candidate ligands by their binding affinitiesEven advances like free energy perturbationFEPand VeraChemandapos s own mining minima free energy method VMremain limited in their ability to consistently provide the accuracy levels neededImportantlyall of these methods have in common a dependency on classical molecular mechanicsMMforce fieldsand even the best force fields for proteins and drug like molecules are not guaranteed to have optimal parameters nor to provide adequate descriptions of chemical interactions involvingfor examplestackingpolarizationcharge transferor metal cationsIn factthe approximations inherent in typical force fields are thought to be a key factor limiting accuracyIn this fasttrack SBIR proposalwe aim to address this key limitation by integrating VeraChemandapos s free energy method VMwith quantum mechanicalQMpotentialsproducing a new software package for QM based protein ligand free energy calculations called PLQM VMThis package will be distinct from other free energy methodssuch as FEPwhich is not readily implemented with QM potentialsSimilarlyalthough QM has been applied to protein ligand systemsexisting methods are limited to focusing on a single conformationwhereas PLQMVMwill integrate existing force field based conformational searching with QM energy and free energy refinementPhase I will provide a first level of QM protein ligand free energy capabilityintegrating VMwith a fast semi empirical QM treatment of the ligand and protein active siteIn Phase IIa capability to allow fast and accurate inclusion of protein atoms beyond the active site will be added through a SEQM polarizable force field methodand a very efficient QM fragmentation scheme will enable energy corrections at higher level QMParallelization on CPUs and GPUs will provide fast enough turnaround to support industry Randamp Dand submission of calculations to both local computer clusters and cloud resources will be supportedThe package will be tested and best practices defined through application to multiple protein targets each with high quality measured affinities for a large series of non covalent inhibitors Finding new drug molecules to treat disease is very difficult and if scientists were able to use computer software programs to reliably predict how strongly drug candidate molecules bind with a particular proteinthis would speed upand make less expensivethe drug discovery processas they would have to make and test a lot fewer molecules in the laboratoryCurrently available software packages are of some helpbut even the best ones do not provide the reliable accuracy scientists really needlargely because they model the interactions between molecules using classical physicswhereas these interactions require quantum mechanics to describe them properlyThis project aims to introduce quantum mechanics based descriptions of molecular interactions into a current state of the art software package for predictions of binding strengthsthereby improving its reliability