Improving performance of screening MM/PBSA in protein-ligand interactions via machine learning

doi:10.1088/1674-1056/ad8ecb

中国物理B ›› 2025, Vol. 34 ›› Issue (1): 18701-018701.doi: 10.1088/1674-1056/ad8ecb

Improving performance of screening MM/PBSA in protein-ligand interactions via machine learning

Yuan-Qiang Chen(陈远强)¹, Yao Xu(徐耀)², Yu-Qiang Ma(马余强)², and Hong-Ming Ding(丁泓铭)^1,†

1 Center for Soft Condensed Matter Physics and Interdisciplinary Research, School of Physical Science and Technology, Soochow University, Suzhou 215006, China;
2 National Laboratory of Solid State Microstructures and Department of Physics, Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210093, China

收稿日期:2024-09-29 修回日期:2024-10-20 接受日期:2024-11-05 发布日期:2025-01-02
通讯作者: Hong-Ming Ding E-mail:dinghm@suda.edu.cn
基金资助:
Project supported by the National Natural Science Foundation of China (Grant Nos. 12222506, 12347102, 12447164, and 12174184).

Improving performance of screening MM/PBSA in protein-ligand interactions via machine learning

Yuan-Qiang Chen(陈远强)¹, Yao Xu(徐耀)², Yu-Qiang Ma(马余强)², and Hong-Ming Ding(丁泓铭)^1,†

1 Center for Soft Condensed Matter Physics and Interdisciplinary Research, School of Physical Science and Technology, Soochow University, Suzhou 215006, China;
2 National Laboratory of Solid State Microstructures and Department of Physics, Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210093, China

Received:2024-09-29 Revised:2024-10-20 Accepted:2024-11-05 Published:2025-01-02
Contact: Hong-Ming Ding E-mail:dinghm@suda.edu.cn
Supported by:
Project supported by the National Natural Science Foundation of China (Grant Nos. 12222506, 12347102, 12447164, and 12174184).

1. 2025-018701-SI.pdf(222KB)

摘要/Abstract

摘要： Accurately estimating protein-ligand binding free energy is crucial for drug design and biophysics, yet remains a challenging task. In this study, we applied the screening molecular mechanics/Poisson-Boltzmann surface area (MM/PBSA) method in combination with various machine learning techniques to compute the binding free energies of protein-ligand interactions. Our results demonstrate that machine learning outperforms direct screening MM/PBSA calculations in predicting protein-ligand binding free energies. Notably, the random forest (RF) method exhibited the best predictive performance, with a Pearson correlation coefficient ($r_{\rm p}$) of 0.702 and a mean absolute error (MAE) of 1.379 kcal/mol. Furthermore, we analyzed feature importance rankings in the gradient boosting (GB), adaptive boosting (AdaBoost), and RF methods, and found that feature selection significantly impacted predictive performance. In particular, molecular weight (MW) and van der Waals (VDW) energies played a decisive role in the prediction. Overall, this study highlights the potential of combining machine learning methods with screening MM/PBSA for accurately predicting binding free energies in biosystems.

关键词: molecular mechanics/Poisson-Boltzmann surface area (MM/PBSA), binding free energy, machine learning, protein-ligand interaction

Abstract: Accurately estimating protein-ligand binding free energy is crucial for drug design and biophysics, yet remains a challenging task. In this study, we applied the screening molecular mechanics/Poisson-Boltzmann surface area (MM/PBSA) method in combination with various machine learning techniques to compute the binding free energies of protein-ligand interactions. Our results demonstrate that machine learning outperforms direct screening MM/PBSA calculations in predicting protein-ligand binding free energies. Notably, the random forest (RF) method exhibited the best predictive performance, with a Pearson correlation coefficient ($r_{\rm p}$) of 0.702 and a mean absolute error (MAE) of 1.379 kcal/mol. Furthermore, we analyzed feature importance rankings in the gradient boosting (GB), adaptive boosting (AdaBoost), and RF methods, and found that feature selection significantly impacted predictive performance. In particular, molecular weight (MW) and van der Waals (VDW) energies played a decisive role in the prediction. Overall, this study highlights the potential of combining machine learning methods with screening MM/PBSA for accurately predicting binding free energies in biosystems.

Key words: molecular mechanics/Poisson-Boltzmann surface area (MM/PBSA), binding free energy, machine learning, protein-ligand interaction

中图分类号: (Molecular dynamics simulation)

87.10.Tf

87.15.-v (Biomolecules: structure and physical properties) 87.15.A- (Theory, modeling, and computer simulation) 87.15.kp (Protein-ligand interactions)

Yuan-Qiang Chen(陈远强), Yao Xu(徐耀), Yu-Qiang Ma(马余强), and Hong-Ming Ding(丁泓铭). Improving performance of screening MM/PBSA in protein-ligand interactions via machine learning[J]. 中国物理B, 2025, 34(1): 18701-018701.

参考文献

[1] Persch E, Dumele O and Diederich F 2015 Angew. Chem. Int. Ed. 54 3290
[2] Tosstorff A, Cole J C, Bartelt R and Kuhn B 2021 ChemMedChem 16 3428
[3] Janin J 1995 Prog. Biophys. Mol. Biol. 64 145
[4] Steinbrecher T and Labahn A 2010 Curr. Med. Chem. 17 767
[5] Acuner Ozbabacan S E, Gursoy A, Keskin O and Nussinov R 2010 Curr. Opin. Drug Discovery Dev. 13 527 PMID: 20812144
[6] Du X, Li Y, Xia Y L, Ai S M, Liang J, Sang P, Ji X L and Liu S Q 2016 Int. J. Mol. Sci. 17 144
[7] Bartas M, Slychko K, Brázda V, Č ervenˇ J, Beaudoin C A, Blundell T L and Pečinka P 2022 Int. J. Mol. Sci. 23 768
[8] Orphanides G and Reinberg D 2002 Cell 108 439
[9] Chaires J B 2008 Annu. Rev. Biophys. 37 135
[10] Yang X, Zhou P, Shen S, Hu Q, Tian C, Xia A, Wang Y, Yang Z, Nan J and Zhou Y 2024 Proc. Natl. Acad. Sci. USA 121 e2401091121
[11] Shivakumar D, Williams J, Wu Y, Damm W, Shelley J and ShermanW 2010 J. Chem. Theory Comput 6 1509
[12] Schreiber G 2002 Curr. Opin. Struct. Biol. 12 41
[13] Hu X and Contini A 2019 J. Chem. Inf. Model. 59 2714
[14] Fu T, Wu X, Xiu Z, Wang J, Yin L and Li G 2013 J. Theor. Comput. Chem. 12 1341003
[15] Wang E, Sun H, Wang J,Wang Z, Liu H, Zhang J Z H and Hou T 2019 Chem. Rev. 119 9478
[16] Wang C, Greene D A, Xiao L, Qi R and Luo R 2018 Front. Mol. Biosci. 4 87
[17] Abel R, Wang L, Harder E D, Berne B J and Friesner R A 2017 Acc. Chem. Res. 50 1625
[18] Wang L, Wu Y, Deng Y, Kim B, Pierce L, Krilov G, Lupyan D, Robinson S, DahlgrenMK, Greenwood J, Romero D L, Masse C, Knight J L, Steinbrecher T, Beuming T, Damm W, Harder E, Sherman W, Brewer M, Wester R, Murcko M, Frye L, Farid R, Lin T, Mobley D L, Jorgensen W L, Berne B J, Friesner R A and Abel R 2015 J. Am. Chem. Soc. 137 2695
[19] Homeyer N and Gohlke H 2012 Mol. Inf. 31 114
[20] Wang E, Weng G, Sun H, Du H, Zhu F, Chen F, Wang Z and Hou T 2019 Phys. Chem. Chem. Phys. 21 18958
[21] Sheng Y J, Yin YW, Ma Y Q and Ding H M 2021 J. Chem. Inf. Model. 61 2454
[22] Ding H M, Yin Y W, Sheng Y J and Ma Y Q 2021 Chin. Phys. Lett. 38 018701
[23] Zhu Y X, Sheng Y J, Ma Y Q and Ding H M 2022 J. Phys. Chem. B 126 1700
[24] Chen Y Q, Sheng Y J, Ding H M and Ma Y Q 2022 Chin. Phys. B 31 048701
[25] Yan Z S, Xu Y, Ding H M and Ma Y Q 2022 Chin. Phys. Lett. 39 108701
[26] Zhang C 2012 Ensemble Machine Learning: Methods and Applications (New York: Springer) pp. 167-184
[27] Chen X, Wang K, Chen J, Wu C, Mao J, Song Y, Liu Y, Shao Z and Pu X 2024 Nat. Commun. 15 8130
[28] Zhang Y, Li S, Meng K and Sun S 2024 J. Chem. Inf. Model. 64 1456
[29] Li J, Fu A and Zhang L 2019 Interdiscip. Sci.: Comput. Life Sci. 11 320
[30] Ballester P J and Mitchell J B 2010 Bioinformatics 26 1169
[31] Sato T, Honma T and Yokoyama S 2010 J. Chem. Inf. Model. 50 170
[32] Deng W, Breneman C and Embrechts M J 2004 J. Chem. Inf. Comput. Sci. 44 699
[33] Staszak M, Staszak K, Wieszczycka K, Bajek A, Roszkowski K and Tylkowski B 2022 Wiley Interdiscip. Rev.: Comput. Mol. Sci. 12 e1568
[34] Ellingson S R, Davis B and Allen J 2020 Biochim. Biophys. Acta, Gen. Subj. 1864 129545
[35] S Heck G, O Pintro V, R Pereira R, Mb Levin N and F De Azevedo W 2017 Curr. Med. Chem. 24 2459
[36] Colwell L J 2018 Curr. Opin. Struct. Biol. 49 123
[37] Bitencourt-Ferreira G and De Azevedo W F 2018 Biophys. Chem. 240 63
[38] Su M, Yang Q, Du Y, Feng G, Liu Z, Li Y and Wang R 2018 J. Chem. Inf. Model. 59 895
[39] Rose P W, Beran B, Bi C, Bluhm W F, Dimitropoulos D, Goodsell D S, Prlić A, Quesada M, Quinn G B and Westbrook J D 2010 Nucleic Acids Res. 39 D392
[40] Burley S K, Berman H M, Christie C, Duarte J M, Feng Z, Westbrook J, Young J and Zardecki C 2018 Protein Sci. 27 316
[41] Abraham M J, Murtola T, Schulz R, Páll S, Smith J C, Hess B and Lindahl E 2015 SoftwareX 1 19
[42] Berendsen H J, Van Der Spoel D and Van Drunen R 1995 Comput. Phys. Commun. 91 43
[43] Van Der Spoel D, Lindahl E, Hess B, Groenhof G, Mark A E and Berendsen H J C 2005 J. Comput. Chem. 26 1701
[44] Case D A, Aktulga H M, Belfon K, Ben-Shalom I, Brozell S R, Cerutti D S, Cheatham Iii T E, Cruzeiro V W D, Darden T A and Duke R E 2020 Amber 2020 (University of California, San Francisco)
[45] Jakalian A, Bush B L, Jack D B and Bayly C I 2000 J. Comput. Chem. 21 132
[46] Jakalian A, Jack D B and Bayly C I 2002 J. Comput. Chem. 23 1623
[47] Wang J, Wolf R M, Caldwell J W, Kollman P A and Case D A 2004 J. Comput. Chem. 25 1157
[48] Maier J A, Martinez C, Kasavajhala K, Wickstrom L, Hauser K E and Simmerling C 2015 J. Chem. Theory Comput. 11 3696
[49] Jorgensen W L and Madura J D 1983 J. Am. Chem. Soc. 105 1407
[50] Hess B, Bekker H, Berendsen H J and Fraaije J G 1997 J. Comput. Chem. 18 1463
[51] Zhou R, Harder E, Xu H and Berne B 2001 J. Chem. Phys. 115 2348
[52] Essmann U, Perera L, Berkowitz M L, Darden T, Lee H and Pedersen L G 1995 J. Chem. Phys. 103 8577
[53] Sun H, Duan L, Chen F, Liu H, Wang Z, Pan P, Zhu F, Zhang J Z and Hou T 2018 Phys. Chem. Chem. Phys. 20 14450
[54] Genheden S and Ryde U 2015 Expert Opin. Drug Discovery 10 449
[55] Rastelli G, Rio A D, Degliesposti G and Sgobba M 2010 J. Comput. Chem. 31 797
[56] Sun H, Li Y, Tian S, Xu L and Hou T 2014 Phys. Chem. Chem. Phys. 16 16719
[57] Chen F, Sun H, Wang J, Zhu F, Liu H, Wang Z, Lei T, Li Y and Hou T 2018 Rna 24 1183
[58] https://jerkwin.github.io/gmxtool
[59] Chen Y Q, Sheng Y J, Ma Y Q and Ding HM2022 Phys. Chem. Chem. Phys. 24 14339
[60] Abraham A, Pedregosa F, Eickenberg M, Gervais P, Mueller A, Kossaifi J, Gramfort A, Thirion B and Varoquaux G 2014 Front. Neuroinform. 8 71792
[61] Genheden S and Ryde U 2011 J. Chem. Theory Comput. 7 3768
[62] He X, Man V H, Ji B, Xie X Q and Wang J 2019 J. Comput.-Aided Mol. Des. 33 105

Improving performance of screening MM/PBSA in protein-ligand interactions via machine learning

Improving performance of screening MM/PBSA in protein-ligand interactions via machine learning

PDF (PC)

赞

补充材料

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	Xuhe Gong(巩旭菏), Ran Li(李然), Ruijuan Xiao(肖睿娟), Tao Zhang(张涛), and Hong Li(李泓). Database of ternary amorphous alloys based on machine learning[J]. 中国物理B, 2025, 34(1): 16101-016101.
[2]	Bohan Zheng(郑博涵), Siyu Zhu(朱思宇), Xingping Zhou(周兴平), and Tong Liu(刘通). Classifying extended, localized and critical states in quasiperiodic lattices via unsupervised learning[J]. 中国物理B, 2025, 34(1): 17103-017103.
[3]	Ming-Hao Wang(王明浩) and Hua Lü(吕桦). Variational data encoding and correlations in quantum-enhanced machine learning[J]. 中国物理B, 2024, 33(9): 90307-090307.
[4]	Rui Xin(辛睿), Yaqi Wang(王亚祺), Ze Fang(房泽), Fengji Zheng(郑凤基), Wen Gao(高雯), Dashi Fu(付大石), Guoqing Shi(史国庆), Jian-Yi Liu(刘建一), and Yongcheng Zhang(张永成). Physics-embedded machine learning search for Sm-doped PMN-PT piezoelectric ceramics with high performance[J]. 中国物理B, 2024, 33(8): 87701-087701.
[5]	Menghui Mao(毛梦辉), Wei Zhou(周唯), Xinhui Li(李新慧), Ran Yang(杨然), Yan-Xiao Gong(龚彦晓), and Shi-Ning Zhu(祝世宁). Machine-learning-assisted efficient reconstruction of the quantum states generated from the Sagnac polarization-entangled photon source[J]. 中国物理B, 2024, 33(8): 80301-080301.
[6]	Bin Liu(刘斌) and Yixi Wang(王译浠). A graph neural network approach to the inverse design for thermal transparency with periodic interparticle system[J]. 中国物理B, 2024, 33(8): 84401-084401.
[7]	Yaqiang Zhang(张亚强), Xuwen Wang(王煦文), Yanan Wang(王雅楠), and Wen Zheng(郑文). Physical information-enhanced graph neural network for predicting phase separation[J]. 中国物理B, 2024, 33(7): 70702-070702.
[8]	Ting Liu(刘婷), Rong-Sheng Han(韩榕生), and Liang Chen(陈亮). Prediction of impurity spectrum function by deep learning algorithm[J]. 中国物理B, 2024, 33(5): 57102-057102.
[9]	Yang Li(李扬), Shenglan Yuan(袁胜兰), Linghongzhi Lu(陆凌宏志), and Xianbin Liu(刘先斌). Computing large deviation prefactors of stochastic dynamical systems based on machine learning[J]. 中国物理B, 2024, 33(4): 40501-040501.
[10]	Jian Zhang(张健), Hao-Chun Zhang(张昊春), Weifeng Li(李伟峰), and Gang Zhang(张刚). Thermal conductivity of GeTe crystals based on machine learning potentials[J]. 中国物理B, 2024, 33(4): 47402-047402.
[11]	Tao Cheng(程涛), Run-Sheng Zhao(赵润盛), Shuang Wang(王爽), Rui Wang(王睿), and Hong-Yang Ma(马鸿洋). Analysis of learnability of a novel hybrid quantum—classical convolutional neural network in image classification[J]. 中国物理B, 2024, 33(4): 40303-040303.
[12]	Yiming Tang(唐一鸣), Zhongyuan Yang(杨中元), Yifei Yao(姚逸飞), Yun Zhou(周运), Yuan Tan(谈圆),Zichao Wang(王子超), Tong Pan(潘瞳), Rui Xiong(熊瑞), Junli Sun(孙俊力), and Guanghong Wei(韦广红). Recent advances in protein conformation sampling by combining machine learning with molecular simulation[J]. 中国物理B, 2024, 33(3): 30701-030701.
[13]	Minyang Zhang(张敏洋), Dong-Xu Chen(陈东旭), Pengxiang Ruan(阮鹏祥), Jun Liu(刘俊), Dong-Zhi Fu(付栋之), Jun-Long Zhao(赵军龙), and Chui-Ping Yang(杨垂平). Deep-learning-assisted optical communication with discretized state space of structured light[J]. 中国物理B, 2024, 33(12): 120304-120304.
[14]	Yizhi Wang(王一志), Xiuhua Cui(崔秀花), Jing Liu(刘静), Qun Jing(井群), Haiming Duan(段海明), and Haibin Cao(曹海宾). Geometries and electronic structures of Zr_nCu(n =2-12) clusters: A joint machine-learning potential density functional theory investigation[J]. 中国物理B, 2024, 33(1): 16109-16109.
[15]	Maosheng Hao(郝茂生) and Pengfei Guan(管鹏飞). An artificial neural network potential for uranium metal at low pressures[J]. 中国物理B, 2023, 32(9): 98401-098401.