<i>In silico</i> binding affinity prediction for metabotropic glutamate receptors using both endpoint free energy methods and a machine learning-based scoring function

Zhai, Jingchen; He, Xibing; Sun, Yuchen; Wan, Zhuoya; Ji, Beihong; Liu, Shuhan; Li, Song; Wang, Junmei

doi:10.1039/d2cp01727j

Cited by 4 publications

(6 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Encouragingly, the prediction results for Set C show even higher prediction accuracy (Figure 6): RMSE = 0.051 g/cm 3 , APE = 2.477%, and R-square = 0.965 for the GAFF descriptor system; RMSE = 0.048 g/cm 3 , APE = 2.150%, and R-square = 0.970 for the RDKit descriptor system. This additional validation can indicate the robustness of our machine learning model for density prediction.…”

Section: Additional Validation On Prediction Of Densitymentioning

confidence: 97%

“…Gaussian Process Regression (GPR) exhibited the most promising performance according to the calculated RMSE and R-square values (Supplementary Table 1), and different GPR sub-type models possess similar performance. Especially, the exponential GPR model ranked the best for the test set (Set B), reporting the RMSE of 0.098 g/cm 3 and the R-square of 0.91. Similarly, for the RDKit descriptors, GPR models with various sub-types also performed the best among all the training algorithms (Supplementary Table 2).…”

Section: Density Predictionmentioning

confidence: 99%

“…2 ∆𝐺 = ∆𝐺 𝑛𝑝 + ∆𝐺 𝑝 (1) The non-polar term can be further decomposed into cavity (cav) and van der Waals dispersion (vdW) terms, and is usually estimated using solvation-accessible surface area (SASA) (Equation 2) with a linear relationship. [2][3][4] On the counterpart, the calculation of polar solvation energy is commonly estimated with the Poisson-Boltzmann (PB) theory or a generalized-Born (GB) model. 5 ∆𝐺 𝑛𝑝 = 𝛾𝑆𝐴𝑆𝐴 + 𝑏 (2) To point out, the calculation of SASA involves the surface tension parameter and dielectric constant, 6,7 and the PB/GB calculation is also affected by dielectric constant of the solute.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Toward Predicting Solubility of Arbitrary Solutes in Arbitrary Solvents. 1: Prediction of Density and Refractive Index Using Machine Learning Algorithms with correlation-group parallel feature analysis

Hu,

Zhai,

et al. 2023

Preprint

View full text Add to dashboard Cite

Density and refractive index (nD) are two important properties related to van der Waals energy of a molecule. Thus, accurate prediction of these two properties has a great value in both molecular mechanics force field development, and solvation free energy and solubility prediction of any arbitrary molecules. In this study, we gathered molecule characteristics information of roughly 5,000 organic compounds for density records and 4000 organic compounds for nD values. Subsequently, the distinct GAFF (General AMBER Force Field) descriptors and RDkit descriptors of the compounds were generated and then applied to train various prediction models with a variety of machine learning algorithms for both properties respectively. As a result, both GAFF and RDkit descriptors yielded various robust models with low average percent errors (APE), low root-mean-square errors (RMSE) and high correlation coefficients R-square, while RDkit showed slightly better performance for predicting both properties. We further optimized top models and conducted parallel feature analysis (PFA) to identify specific features in each descriptor which outstandingly contributed to model robustness. The final model RMSE is 0.071 g/cm3 for density prediction and 0.014 for nD prediction, the APE value is as low as 2.845% for density and 0.531% for nD, and R-square is 0.950 for density and 0.954 for nD. Note that the performance of our prediction models for both density and nD significantly outperforms all currently published studies, especially for those with a dataset containing more than 200 records. The successful prediction of the two key molecular properties paves the road towards accurately predicting solubility of an arbitrary solute in an arbitrary solvent, an endeavor not only facilitates pharmaceutical industry to develop better drug candidates, but also increases efficiency regarding overall wet lab work. Key predictors which contribute most to a specific model or model function were identified using both Shapley analysis and correlation-group parallel feature analysis (CG-PFA).

show abstract

Section: Additional Validation On Prediction Of Densitymentioning

confidence: 97%

Section: Density Predictionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Toward Predicting Solubility of Arbitrary Solutes in Arbitrary Solvents. 1: Prediction of Density and Refractive Index Using Machine Learning Algorithms with correlation-group parallel feature analysis

Hu,

Zhai,

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…End-point free energy techniques as pivotal tools in drug discovery are commonly recognized as regimes with intermediate accuracy and efficiency lying between molecular docking and (alchemical) free energy calculation. In protein–ligand and protein–protein interactions, such a computational ladder is believed to be solid, with a huge library of case reports supporting its usage [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 ]. However, such ‘common sense’ seems broken in host–guest complexes, although the latter are often recognized as prototypical models of biomacromolecular systems.…”

Section: Introductionmentioning

confidence: 99%

Comprehensive Evaluation of End-Point Free Energy Techniques in Carboxylated-Pillar[6]arene Host–Guest Binding: III. Force-Field Comparison, Three-Trajectory Realization and Further Dielectric Augmentation

et al. 2023

View full text Add to dashboard Cite

Host–guest binding, despite the relatively simple structural and chemical features of individual components, still poses a challenge in computational modelling. The extreme underperformance of standard end-point methods in host–guest binding makes them practically useless. In the current work, we explore a potentially promising modification of the three-trajectory realization. The alteration couples the binding-induced structural reorganization into free energy estimation and suffers from dramatic fluctuations in internal energies in protein–ligand situations. Fortunately, the relatively small size of host–guest systems minimizes the magnitude of internal fluctuations and makes the three-trajectory realization practically suitable. Due to the incorporation of intra-molecular interactions in free energy estimation, a strong dependence on the force field parameters could be incurred. Thus, a term-specific investigation of transferable GAFF derivatives is presented, and noticeable differences in many aspects are identified between commonly applied GAFF and GAFF2. These force-field differences lead to different dynamic behaviors of the macrocyclic host, which ultimately would influence the end-point sampling and binding thermodynamics. Therefore, the three-trajectory end-point free energy calculations are performed with both GAFF versions. Additionally, due to the noticeable differences between host dynamics under GAFF and GAFF2, we add additional benchmarks of the single-trajectory end-point calculations. When only the ranks of binding affinities are pursued, the three-trajectory realization performs very well, comparable to and even better than the regressed PBSA_E scoring function and the dielectric constant-variable regime. With the GAFF parameter set, the TIP3P water in explicit solvent sampling and either PB or GB implicit solvent model in free energy estimation, the predictive power of the three-trajectory realization in ranking calculations surpasses all existing end-point methods on this dataset. We further combine the three-trajectory realization with another promising modified end-point regime of varying the interior dielectric constant. The combined regime does not incur sizable improvements for ranks and deviations from experiment exhibit non-monotonic variations.

show abstract

“…In protein-ligand and protein-protein interactions, such a computational ladder is believed to be solid, with a huge library of case reports supporting its usage. [1][2][3][4][5][6][7][8] However, such 'common sense' seems broken in host-guest complexes, although the latter (i.e., host-guest complexes) are often recognized as prototypical models of biomacromolecular systems. Specifically, although rigorous free energy simulations still achieve better performance than end-point methods, [9][10][11] docking calculations could produce rather small deviations from experimental reference compared with other costly free energy techniques.…”

Section: Introductionmentioning

confidence: 99%

Comprehensive Evaluation of End-Point Free Energy Techniques in Carboxylated-Pillar[6]arene Host-guest Binding: III. Force-Field Comparison, Three-Trajectory Realization and Further Dielectric Augmentation

Liu

Zheng

Qin

et al. 2023

Preprint

View full text Add to dashboard Cite

Host-guest binding, despite the relatively simple structural and chemical features of individual components, still poses a challenge in computational modelling. The problems lie in both the accuracy of the employed Hamiltonian (often fixed-charge force fields) and the exhaustiveness of conformational sampling. End-point free energy calculations as fast alternatives to rigorous but costly methods are widely applied in virtual screening in protein-ligand and host-guest systems. However, the extreme underperformance of standard end-point methods makes them practically useless. Modifications of the end-point procedure could bring these methods back to the pool of usable tools, e.g., regression considered in our previous work. In the current work, we explore a potentially promising modification, the three-trajectory realization of the end-point simulation protocol. The alteration couples the binding-induced structural reorganization into free energy estimation and suffers from dramatic fluctuations of internal energies in protein-ligand situations. Fortunately, the relatively small size of host-guest systems minimizes the magnitude of internal fluctuations and makes the three-trajectory realization practically suitable. Due to the incorporation of intra-molecular interactions in free energy estimation, a strong dependence on the force field parameters could be incurred. Thus, a term-specific investigation of transferable GAFF derivatives is presented, and noticeable differences in many aspects are identified between commonly applied GAFF and GAFF2. These force-field differences lead to different dynamic behaviors of the macrocyclic host, which ultimately would influence the end-point sampling and binding thermodynamics. Therefore, the three-trajectory end-point free energy calculations are performed with both GAFF versions to investigate the force-field dependent behavior of computed binding affinities. Also, due to the noticeable differences between host dynamics under GAFF and GAFF2, we add additional benchmarks of the single-trajectory end-point calculations. Numerical results suggest that the single-trajectory realization, regardless of the GAFF version, is still not useful in host-guest binding, although the prediction quality of the GAFF2 parameter set is slightly better than GAFF. As for the three-trajectory realization, the absolute values of computed binding thermodynamics exhibit pronounced force-field-dependent behaviors, which are less significant for ranking information. When only the ranks of binding affinities are pursued, the three-trajectory realization performs very well, comparable to and even better than the regressed PBSA_E scoring function and the dielectric-constant-variable regime. With the GAFF parameter set, the TIP3P water in explicit-solvent sampling and either PB or GB implicit-solvent model in free energy estimation, the predictive power of the three-trajectory realization in ranking calculations surpasses all existing end-point methods on this dataset. We further combine the three-trajectory realization with another promising modified end-point regime of varying the interior dielectric constant. The predicted binding affinities exhibit monotonic responses to the variation of the internal dielectric constant, but the deviations from experiment exhibit non-monotonic variations, which are related to the systematic overestimation of the binding strength under the original three-trajectory realization. By contrast, the combined regime does not incur sizable improvements for ranks, although for most systems the dielectric constant 2 seems to be the best option.

show abstract

In silico binding affinity prediction for metabotropic glutamate receptors using both endpoint free energy methods and a machine learning-based scoring function

Abstract: The metabotropic glutamate receptors (mGluRs) play an important role in regulating glutamate signal pathways, which involves in neuropathy and periphery homeostasis. The mGluR4, which belongs to Group III mGluRs, is...

Cited by 4 publications

References 49 publications

Toward Predicting Solubility of Arbitrary Solutes in Arbitrary Solvents. 1: Prediction of Density and Refractive Index Using Machine Learning Algorithms with correlation-group parallel feature analysis

Toward Predicting Solubility of Arbitrary Solutes in Arbitrary Solvents. 1: Prediction of Density and Refractive Index Using Machine Learning Algorithms with correlation-group parallel feature analysis

Comprehensive Evaluation of End-Point Free Energy Techniques in Carboxylated-Pillar[6]arene Host–Guest Binding: III. Force-Field Comparison, Three-Trajectory Realization and Further Dielectric Augmentation

Comprehensive Evaluation of End-Point Free Energy Techniques in Carboxylated-Pillar[6]arene Host-guest Binding: III. Force-Field Comparison, Three-Trajectory Realization and Further Dielectric Augmentation

Contact Info

Product

Resources

About