Evaluation of several two-step scoring functions based on linear interaction energy, effective ligand size, and empirical pair potentials for prediction of protein-ligand binding geometry and free energy

The performances of several two-step scoring approaches for molecular docking were assessed for their ability to predict binding geometries and free energies. Two new scoring functions designed for "step 2 discrimination" were proposed and compared to our CHARMM implementation of the linear interaction energy (LIE) approach using the Generalized-Born with Molecular Volume (GBMV) implicit solvation model. A scoring function S1 was proposed by considering only "interacting" ligand atoms as the "effective size" of the ligand and extended to an empirical regression-based pair potential S2. The S1 and S2 scoring schemes were trained and 5-fold cross-validated on a diverse set of 259 protein-ligand complexes from the Ligand Protein Database (LPDB). The regression-based parameters for S1 and S2 also demonstrated reasonable transferability in the CSARdock 2010 benchmark using a new data set (NRC HiQ) of diverse protein-ligand complexes. The ability of the scoring functions to accurately predict ligand geometry was evaluated by calculating the discriminative power (DP) of the scoring functions to identify native poses. The parameters for the LIE scoring function with the optimal discriminative power (DP) for geometry (step 1 discrimination) were found to be very similar to the best-fit parameters for binding free energy over a large number of protein-ligand complexes (step 2 discrimination). Reasonable performance of the scoring functions in enrichment of active compounds in four different protein target classes established that the parameters for S1 and S2 provided reasonable accuracy and transferability. Additional analysis was performed to definitively separate scoring function performance from molecular weight effects. This analysis included the prediction of ligand binding efficiencies for a subset of the CSARdock NRC HiQ data set where the number of ligand heavy atoms ranged from 17 to 35. This range of ligand heavy atoms is where improved accuracy of predicted ligand efficiencies is most relevant to real-world drug design efforts.     

Robust scoring functions for protein-ligand interactions with quantum chemical charge models

Ordinary least-squares (OLS) regression has been used widely for constructing the scoring functions for protein-ligand interactions. However, OLS is very sensitive to the existence of outliers, and models constructed using it are easily affected by the outliers or even the choice of the data set. On the other hand, determination of atomic charges is regarded as of central importance, because the electrostatic interaction is known to be a key contributing factor for biomolecular association. In the development of the AutoDock4 scoring function, only OLS was conducted, and the simple Gasteiger method was adopted. It is therefore of considerable interest to see whether more rigorous charge models could improve the statistical performance of the AutoDock4 scoring function. In this study, we have employed two well-established quantum chemical approaches, namely the restrained electrostatic potential (RESP) and the Austin-model 1-bond charge correction (AM1-BCC) methods, to obtain atomic partial charges, and we have compared how different charge models affect the performance of AutoDock4 scoring functions. In combination with robust regression analysis and outlier exclusion, our new protein-ligand free energy regression model with AM1-BCC charges for ligands and Amber99SB charges for proteins achieve lowest root-mean-squared error of 1.637 kcal/mol for the training set of 147 complexes and 2.176 kcal/mol for the external test set of 1427 complexes. The assessment for binding pose prediction with the 100 external decoy sets indicates very high success rate of 87% with the criteria of predicted root-mean-squared deviation of less than 2 ?. The success rates and statistical performance of our robust scoring functions are only weakly class-dependent (hydrophobic, hydrophilic, or mixed).     

基于支持向量学习机方法的人体小肠吸收药物活性的预测

为了预测分子在人体小肠中的吸收,本文计算了表征分子的电子、拓扑、几何结构、分子形状等特征的102个分子描述符,用遗传算法变量选择方法使描述符减少到47个。体系共包含了230个化合物分子,69个不能被吸收(mA-),161个可以被吸收(HIA )。对建立的SVM模型,用5重交叉验证和独立测试集进行验证,预测正确率分别达到79．1％和77．1％,结果具有较好的一致性。在模型验证中,通过聚类分析方法组合训练集和测试集,保证了模型的稳定性,提高了建模效率。     

CSAR scoring challenge reveals the need for new concepts in estimating protein-ligand binding affinity

The dG prediction accuracy by the Lead Finder docking software on the CSAR test set was characterized by R(2)=0.62 and rmsd=1.93 kcal/mol, and the method of preparation of the full-atom structures of the test set did not significantly affect the resulting accuracy of predictions. The primary factors determining the correlation between the predicted and experimental values were the van der Waals interactions and solvation effects. Those two factors alone accounted for R(2)=0.50. The other factors that affected the accuracy of predictions, listed in the order of decreasing importance, were the change of ligand's internal energy upon binding with protein, the electrostatic interactions, and the hydrogen bonds. It appears that those latter factors contributed to the independence of the prediction results from the method of full-atom structure preparation. Then, we turned our attention to the other factors that could potentially improve the scoring function in order to raise the accuracy of the dG prediction. It turned out that the ligand-centric factors, including Mw, cLogP, PSA, etc. or protein-centric factors, such as the functional class of protein, did not improve the prediction accuracy. Following that, we explored if the weak molecular interactions such as X-H...Ar, X-H...Hal, CO...Hal, C-H...X, stacking and π-cationic interactions (where X is N or O), that are generally of interest to the medicinal chemists despite their lack of proper molecular mechanical parametrization, could improve dG prediction. Our analysis revealed that out of these new interactions only CO...Hal is statistically significant for dG predictions using Lead FInder scoring function. Accounting for the CO...Hal interaction resulted in the reduction of the rmsd from 2.19 to 0.69 kcal/mol for the corresponding structures. The other weak interaction factors were not statistically significant and therefore irrelevant to the accuracy of dG prediction. On the basis of our findings from our participation in the CSAR scoring challenge we conclude that a significant increase of accuracy predictions necessitates breakthrough scoring approaches. We anticipate that the explicit accounting for water molecules, protein flexibility, and a more thermodynamically accurate method of dG calculation rather than single point energy calculation may lead to such breakthroughs.