Machine learning on ligand-residue interaction profiles to significantly improve binding affinity prediction.

Structure-based virtual screenings (SBVSs) play an important role in drug discovery projects. However, it is still a challenge to accurately predict the binding affinity of an arbitrary molecule binds to a drug target and prioritize top ligands from an SBVS. In this study, we developed a novel method, using ligand-residue interaction profiles (IPs) to construct machine learning (ML)-based prediction models, to significantly improve the screening performance in SBVSs. Such a kind of the prediction model is called an IP scoring function (IP-SF). We systematically investigated how to improve the performance of IP-SFs from many perspectives, including the sampling methods before interaction energy calculation and different ML algorithms. Using six drug targets with each having hundreds of known ligands, we conducted a critical evaluation on the developed IP-SFs. The IP-SFs employing a gradient boosting decision tree (GBDT) algorithm in conjunction with the MIN + GB simulation protocol achieved the best overall performance. Its scoring power, ranking power and screening power significantly outperformed the Glide SF. First, compared with Glide, the average values of mean absolute error and root mean square error of GBDT/MIN + GB decreased about 38 and 36%, respectively. Second, the mean values of squared correlation coefficient and predictive index increased about 225 and 73%, respectively. Third, more encouragingly, the average value of the areas under the curve of receiver operating characteristic for six targets by GBDT, 0.87, is significantly better than that by Glide, which is only 0.71. Thus, we expected IP-SFs to have broad and promising applications in SBVSs.

New application of in silico methods in identifying mechanisms of action and key components of anti-cancer herbal formulation YIV-906 (PHY906).

YIV-906 (formally PHY906, KD018) is a four-herb formulation that is currently being developed to improve the therapeutic index and ameliorate the side effects of many chemotherapeutic drugs including sorafenib, irinotecan, and capecitabine. However, as a promising anti-cancer adjuvant, the molecular mechanism of action of YIV-906 remains unrevealed due to its multi-component and multi-target features. Since YIV-906 has been shown to induce apoptosis and autophagy in cancer cells through modulating the negative regulators of ERK1/2, namely DUSPs, it is of great interest to elucidate the key components that cause the therapeutic effect of YIV-906. In this work, we investigated the mechanism of YIV-906 inhibiting DUSPs, using a broad spectrum of molecular modelling techniques, including molecular docking, molecular dynamics (MD) simulations, and binding free energy calculations. In total, MD simulations and binding free energy calculations were performed for 99 DUSP-ligand complexes. We found that some herbal components or their metabolites could inhibit DUSPs. Based on the docking scores and binding free energies, the sulfation and glucuronidation metabolites of the S ingredient in YIV-906 play a leading role in inhibiting DUSPs, although several original herbal chemicals with carboxyl groups from the P and Z ingredients also make contributions to this inhibitory effect. It is not a surprise that the electrostatic interaction plays the dominant role in the ligand binding process, given the fact that several charged residues reside in the binding pockets of DUSPs. Our MD simulation results demonstrate that the sulfate moieties and carboxyl moieties of the advantageous ligands from YIV-906 can occupy the enzymes' catalytic sites, mimicking the endogenous phosphate substrates of DUSPs. As such, the ligand binding can inhibit the association of DUSPs and ERK1/2, which in turn reduces the dephosphorylation of ERK1/2 and causes cell cycle arrest in the tumor. Our modelling study provides useful insights into the rational design of highly potent anti-cancer drugs targeting DUSPs. Finally, we have demonstrated that multi-scale molecular modelling techniques are able to elucidate molecular mechanisms involving complex molecular systems.