English

Machine-Learning Model for Predicting the Rate Constant of ProteinLigand Dissociation

• Su Minyi ^1,2,†, ,
• Liu Huisi ^3,†, ,
• Lin Haixia ^{3, ,} ,
• Wang Renxiao ^{1,2, ,}

• 1.

  State Key Laboratory of Bioorganic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, Shanghai 200032, P. R. China

• 2.

  University of Chinese Academy of Sciences, Beijing 100049, P. R. China

• 3.

  Department of Chemistry, College of Sciences, Shanghai University, Shanghai 200444, P. R. China

Corresponding author: Email: haixialin@staff.shu.edu.cn (H.L.); Emails: wangrx@mail.sioc.ac.cn (R.W.)

• Received Date: 01 July 2019
  Accepted Date: 30 August 2019
  Revised Date: 06 August 2019
  Available Online: 15 January 2020
• Fund Project:

  This project was supported by the National Key Research Program of Ministry of Science and Technology of China (2016YFA0502302), National Natural Science Foundation of China (81725022, 81430083, 21661162003, 21673276, 21472227, 21472226), and Strategic Priority Research Program of Chinese Academy of Sciences (XDB20000000)

Abstract: An increasing number of recent studies have shown that the binding kinetics of a drug molecule to its target correlates strongly with its efficacy in vivo. Therefore, ligand optimization oriented to improved binding kinetics provides new ideas for rational drug design. Currently, ligand binding kinetics is modeled mainly through extensive molecular dynamics simulations, which limits its application to real-world problems. The present study aimed at obtaining a general-purpose quantitative structure-kinetics relationship (QSKR) model for predicting the dissociation rate constant (k_off) of a ligand based on its complex structure. This type of model is expected to be suitable for high-throughput tasks in structure-based drug design. We collected the experimentally measured k_off values for 406 ligand molecules from literature, and then constructed a three-dimensional structural model for each protein-ligand complex through molecular modeling. A training set was compiled using 60% of those complexes while the remaining 40% were assigned to two test sets. Based on distance-dependent protein-ligand atom pair descriptors, a random forest algorithm was adopted to derive a QSKR model. Various random forest models were then generated based on the descriptor sets obtained under different conditions, such as distance cutoff, bin width, and feature selection criteria. The cross-validation results of those models were then examined. It was observed that the optimal model was obtained when the distance cutoff was 15 Å (1 Å = 0.1 nm), the bin width was 3 Å, and feature selection variance level was 2. The final QSKR model produced correlation coefficients around 0.62 on the two independent test sets. This level of accuracy is at least comparable to that of the predictive models described in literature, which are typically computationally much more expensive. Our study attempts to address the issue of predicting k_off values in drug design. We hope that it can provide inspiration for further studies by other researchers.

Key words:
• Dissociation rate constant
•  /  Ligand binding kinetics
•  /  Random forest model
•  /  Protein-ligand interaction
•  /  Structure-based drug design

• 1. [1]

     Copeland, R. A.; Pompliano, D. L.; Meek, T. D. Nat. Rev. Drug Discov. 2006, 5, 730. doi: 10.1038/nrd2082

  2. [2]

     Tummino, P. J.; Copeland, R. A. Biochemistry 2008, 47, 5481. doi: 10.1021/bi8002023

  3. [3]

     Schuetz, D. A.; Arnout de Witte, W. E.; Wong, Y. C.; Knasmueller, B.; Richter, L.; Kokh, D. B.; Sadiq, S. K.; Bosma, R.; Nederpelt, I.; Heitman, L. H.; et al. Drug Discov. Today 2017, 22, 896. doi: 10.1016/j.drudis.2017.02.002

  4. [4]

     Pan, A. C.; Borhani, D. W.; Dror, R. O.; Shaw, D. E. Drug Discov. Today 2013, 18, 667. doi: 10.1016/j.drudis.2013.02.007

  5. [5]

     Guo, D.; Mulder, K. T.; Ijzerman, A. P.; Heitman, L. H. Br. J. Pharmacol. 2012, 166, 1846. doi: 10.1111/j.1476-5381.2012.01897.x

  6. [6]

     Folmer, R. H. A. Drug Discov. Today 2018, 23, 12. doi: 10.1016/j.drudis.2017.07.016

  7. [7]

     Bruce, N. J.; Ganotra, G. K.; Kokh, D. B.; Sadiq, S. K.; Wade, R. C. Curr. Opin. Struct. Biol. 2018, 49, 1. doi: 10.1016/j.sbi.2017.10.001

  8. [8]

     Qu, S. J.; Huang, S. H.; Pan, X. C.; Yang, L.; Mei, H. J. Chem. Inf. Model. 2016, 56, 2061. doi: 10.1021/acs.jcim.6b00326

  9. [9]

     Ganotra, G. K.; Wade, R. C. ACS Med. Chem. Lett. 2018, 9, 1134. doi: 10.1021/acsmedchemlett.8b00397

  10. [10]

      Liu, Z. H; Li, Y.; Han, L.; Li, J.; Liu, J.; Zhao, Z. X.; Nie, W.; Liu, Y. C; Wang, R. X. Bioinformatics 2015, 31, 405. doi: 10.1093/bioinformatics/btu626

  11. [11]

      Li, W.; Godzik, A. Bioinformatics 2006, 22, 1658. doi: 10.1093/bioinformatics/btl158

  12. [12]

      Fu, L.; Niu, B.; Zhu, Z.; Wu, S.; Li, W. Bioinformatics 2012, 28, 3150. doi: 10.1093/bioinformatics/bts565

  13. [13]

      Royston, J. P. Appl. Stat. 1982, 31, 115. doi: 10.2307/2347973

  14. [14]

      Royston, J. P. Appl. Stat. 1982, 31, 176. doi: 10.2307/2347986

  15. [15]

      Rogers, D.; Hahn, M. J. Chem. Inf. Model. 2010, 50, 742. doi: 10.1021/ci100050t

  16. [16]

      Bietz, S.; Urbaczek, S.; Schulz, B.; Rarey, M. J. Cheminform. 2014, 6, 1. doi: 10.1186/1758-2946-6-12

  17. [17]

      Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry, J. K.; et al. J. Med. Chem. 2004, 47, 1739. doi: 10.1021/jm0306430

  18. [18]

      Halgren, T. A.; Murphy, R. B.; Friesner, R. A.; Beard, H. S.; Frye, L. L.; Pollard, W. T.; Banks, J. L. J. Med. Chem. 2004, 47, 1750. doi: 10.1021/jm030644s

  19. [19]

      Friesner, R. A.; Murphy, R. B.; Repasky, M. P.; Frye, L. L.; Greenwood, J. R.; Halgren, T. A.; Sanschagrin, P. C.; Mainz, D. T. J. Med. Chem. 2006, 49, 6177. doi: 10.1021/jm051256o

  20. [20]

      Qiu, D.; Shenkin, P. S.; Hollinger, F. P.; Still, W. C. J. Phys. Chem. A 1997, 101, 3005. doi: 10.1021/jp961992r

  21. [21]

      Case, D. A.; Babin, V.; Berryman, J. T.; Betz, R. M.; Cai, Q.; Cerutti, D. S.; Cheatham, T. E., Ⅲ; Darden, T. A.; Duke, R. E.; Gohlke, H.; Goetz, A. W.; et al. AMBER 2014; University of California: San Francisco, CA, USA, 2014.

  22. [22]

      Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H.; et al. Gaussian 09, Revision A.02; Gaussian, Inc.: Wallingford, CT, USA, 2016.

  23. [23]

      Cieplak, P.; Cornell, W. D.; Bayly, C.; Kollman, P. A. J. Comput. Chem. 1995, 16, 1357. doi: 10.1002/jcc.540161106

  24. [24]

      Maier, J. A.; Martinez, C.; Kasavajhala, K.; Wickstrom, L.; Hauser, K. E.; Simmerling, C. J. Chem. Theory Comput. 2015, 11, 3696. doi: 10.1021/acs.jctc.5b00255

  25. [25]

      Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D. A. J. Comput. Chem. 2004, 25, 1157. doi: 10.1002/jcc.20035

  26. [26]

      Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L. J. Chem. Phys. 1983, 79, 926. doi: 10.1063/1.445869

  27. [27]

      Ryckaert, J. P.; Ciccotti, G.; Berendsen, H. J. C. J. Chem. Phys. 1977, 23, 327. doi: 10.1.1.399.6868

  28. [28]

      Roe, D. R.; Cheatham, T. E., Ⅲ. J. Chem. Theory Comput. 2013, 9, 3084. doi: 10.1021/ct400341p

  29. [29]

      Rudling, A.; Orro, A.; Carlsson, J. J. Chem. Inf. Model. 2018, 58, 350. doi: 10.1021/acs.jcim.7b00520

  30. [30]

      Ballester, P. J.; Schreyer, A.; Blundell, T. L. J. Chem. Inf. Model. 2014, 54, 944. doi: 10.1021/ci500091r

  31. [31]

      Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Dubourg, V.; et al. J. Mach. Learn. Res. 2011, 12, 2825.

  32. [32]

      Kennard, R. W.; Stone, L. A. Technometrics 1969, 11, 137. doi: 10.1080/00401706.1969.10490666

  33. [33]

      Martin, T. M.; Harten, P.; Young, D. M.; Muratov, E. N.; Golbraikh, A.; Zhu, H.; Tropsha, A. J. Chem. Inf. Model. 2012, 52, 2570. doi: 10.1021/ci300338w

•