基于小数据集的机器学习预测酰胺键合成转化率

引用本文: 李兴海, 吴志森, 张利静, 陶胜洋. 基于小数据集的机器学习预测酰胺键合成转化率[J]. 物理化学学报, 2025, 41(2): 100010. doi: 10.3866/PKU.WHXB202309041 shu

Citation: Xinghai Li, Zhisen Wu, Lijing Zhang, Shengyang Tao. Machine Learning Enables the Prediction of Amide Bond Synthesis Based on Small Datasets[J]. Acta Physico-Chimica Sinica, 2025, 41(2): 100010. doi: 10.3866/PKU.WHXB202309041 shu

基于小数据集的机器学习预测酰胺键合成转化率

大连理工大学化学学院, 精细化工国家重点实验室, 智能材料化工前沿科学中心, 大连市智能化学重点实验室, 辽宁大连 116024

通讯作者: 张利静, E-mail: zhanglj@dlut.edu.cn; 陶胜洋, E-mail: taosy@dlut.edu.cn

基金项目:
国家自然科学基金(22072011,22372025,22211530456)，中央高校基本科研业务费(DUT22LAB607,DUT22QN226)和中国航空研究院1912项目资助

摘要: 机器学习(ML)在分子合成领域显示了重要的应用前景。然而，准确的机器学习预测依赖于大量实验数据，而通过传统实验方法获得成千上万的实验数据仍然是一个巨大的挑战。因此，基于小数据集得到可接受的预测模型是目前该领域亟待解决的重要问题。本研究通过构建1152个反应数据，利用大量有化学意义的特征描述符，通过多维数据分析获得了有效的预测结果，证明了基于小数据集的机器学习算法可以可靠地预测酰胺键合成反应的转化率。研究比较了6种机器学习算法的预测精度，其中随机森林表现出卓越的预测性能(R² > 0.95)。同时，在预测未知芳胺分子的转化率时，研究发现在训练集中加入少量未知分子的相关反应数据，即使数据集较小，也能显著提升对未知分子转化率的预测准确性，揭示了一种利用小数据集得到较好预测结果的方法。本研究为小数据集下的机器学习辅助化学合成研究提供了参考价值。不久的将来，机器学习将有力地推动有机合成化学的智能化发展。

关键词:

English

Machine Learning Enables the Prediction of Amide Bond Synthesis Based on Small Datasets

School of Chemistry, State Key Laboratory of Fine Chemicals, Frontier Science Center for Smart Materials, Dalian Key Laboratory of Intelligent Chemistry, Dalian University of Technology, Dalian 116024, Liaoning Province, China

Corresponding author: Lijing Zhang, ; Shengyang Tao,

Received Date: 27 September 2023
Accepted Date: 30 November 2023
Revised Date: 29 November 2023
Fund Project:
The project was supported by the National Natural Science Foundation of China (22072011, 22372025, 22211530456), the Fundamental Research Funds for the Central Universities (DUT22LAB607, DUT22QN226), and Project 1912 Funds provided by the Chinese Aeronautical Establishment.

Abstract: Machine learning (ML) is progressively revealing notable advantages in chemical synthesis. However, the limited output of experimental data from traditional methods poses a bottleneck, impeding the widespread adoption of machine learning. Data from literature often leads to overly optimistic predictions, and obtaining thousands of experimental data points through experiments remains a substantial challenge. Using a small dataset of experimental data, we illustrated that machine learning algorithms can reliably predict the conversion rate of amide bond synthesis. We gathered hundreds of experimental data points for 9 aromatic amines and 12 organic acids using various coupling reagents and solvents in a 96-well plate high-throughput experimental setup. Subsequently, we derived 76 feature molecular descriptors from quantum chemical calculations and utilized them as inputs for training the machine learning model. Despite the inherent limitation of low data volume, the random forest algorithm demonstrated outstanding predictive performance (R² > 0.95). Through comprehensive analysis of the reaction process employing importance analysis, shapley additive explanations (SHAP), and accumulated local effects (ALE) methods, we delved into the important factors influencing the reaction conversion rate. In predicting the conversion rate of unknown aromatic amine molecules, we discovered that incorporating a small amount of unknown molecule-related reaction data into the training set effectively enhances the model’s predictive performance, even with a small dataset. By comparing models trained on different molecular descriptors such as density functional theory (DFT) and one-hot encoding, we validated the efficacy of adjusting the training set to improve prediction results. This study utilized a multitude of chemically meaningful feature descriptors and achieved more effective prediction results through multidimensional data analysis, offering valuable insights for machine learning-assisted chemical synthesis research in small datasets. In the near future, machine learning is poised to drive the intelligent development of organic chemistry.

Key words:

1. [1]
  (1) Jordan, M. I.; Mitchell, T. M. Science 2015, 349, 255. doi: 10.1126/science.aaa8415(1) Jordan, M. I.; Mitchell, T. M. Science 2015, 349, 255. doi: 10.1126/science.aaa8415
2. [2]
  (2) Young, T.; Hazarika, D.; Poria, S.; Cambria, E. IEEE Comput. Intell. Mag. 2018, 13, 55. doi: 10.1109/mci.2018.2840738(2) Young, T.; Hazarika, D.; Poria, S.; Cambria, E. IEEE Comput. Intell. Mag. 2018, 13, 55. doi: 10.1109/mci.2018.2840738
3. [3]
  (3) Myszczynska, M. A.; Ojamies, P. N.; Lacoste, A. M. B.; Neil, D.; Saffari, A.; Mead, R.; Hautbergue, G. M.; Holbrook, J. D.; Ferraiuolo, L. Nat. Rev. Neurol. 2020, 16, 440. doi: 10.1038/s41582-020-0377-8(3) Myszczynska, M. A.; Ojamies, P. N.; Lacoste, A. M. B.; Neil, D.; Saffari, A.; Mead, R.; Hautbergue, G. M.; Holbrook, J. D.; Ferraiuolo, L. Nat. Rev. Neurol. 2020, 16, 440. doi: 10.1038/s41582-020-0377-8
4. [4]
  (4) Ranjan, R.; Sankaranarayanan, S.; Bansal, A.; Bodla, N.; Chen, J. C.; Patel, V. M.; Castillo, C. D.; Chellappa, R. IEEE Signal Process. Mag. 2018, 35, 66. doi: 10.1109/msp.2017.2764116(4) Ranjan, R.; Sankaranarayanan, S.; Bansal, A.; Bodla, N.; Chen, J. C.; Patel, V. M.; Castillo, C. D.; Chellappa, R. IEEE Signal Process. Mag. 2018, 35, 66. doi: 10.1109/msp.2017.2764116
5. [5]
  (5) Segler, M. H. S.; Waller, M. P. Chem.-Eur. J. 2017, 23, 5966. doi: 10.1002/chem.201605499(5) Segler, M. H. S.; Waller, M. P. Chem.-Eur. J. 2017, 23, 5966. doi: 10.1002/chem.201605499
6. [6]
  (6) Shen, Y.; Borowski, J. E.; Hardy, M. A.; Sarpong, R.; Doyle, A. G.; Cernak, T. Nat. Rev. Method. Prim. 2021, 1, 1. doi: 10.1038/s43586-021-00022-5(6) Shen, Y.; Borowski, J. E.; Hardy, M. A.; Sarpong, R.; Doyle, A. G.; Cernak, T. Nat. Rev. Method. Prim. 2021, 1, 1. doi: 10.1038/s43586-021-00022-5
7. [7]
  (7) Brockherde, F.; Vogt, L.; Li, L.; Tuckerman, M. E.; Burke, K.; Müller, K. R. Nat. Commun. 2017, 8, 1. doi: 10.1038/s41467-017-00839-3(7) Brockherde, F.; Vogt, L.; Li, L.; Tuckerman, M. E.; Burke, K.; Müller, K. R. Nat. Commun. 2017, 8, 1. doi: 10.1038/s41467-017-00839-3
8. [8]
  (8) Dara, S.; Dhamercherla, S.; Jadav, S. S.; Babu, C. M.; Ahsan, M. J. Artif. Intell. Rev. 2022, 55, 1947. doi: 10.1007/s10462-021-10058-4(8) Dara, S.; Dhamercherla, S.; Jadav, S. S.; Babu, C. M.; Ahsan, M. J. Artif. Intell. Rev. 2022, 55, 1947. doi: 10.1007/s10462-021-10058-4
9. [9]
  (9) Ahneman, D. T.; Estrada, J. G.; Lin, S. S.; Dreher, S. D.; Doyle, A. G. Science 2018, 360, 186. doi: 10.1126/science.aar5169(9) Ahneman, D. T.; Estrada, J. G.; Lin, S. S.; Dreher, S. D.; Doyle, A. G. Science 2018, 360, 186. doi: 10.1126/science.aar5169
10. [10]
  (10) Raccuglia, P.; Elbert, K. C.; Adler, P. D.; Falk, C.; Wenny, M. B.; Mollo, A.; Zeller, M.; Friedler, S. A.; Schrier, J.; Norquist, A. J. Nature 2016, 533, 73. doi: 10.1038/nature17439(10) Raccuglia, P.; Elbert, K. C.; Adler, P. D.; Falk, C.; Wenny, M. B.; Mollo, A.; Zeller, M.; Friedler, S. A.; Schrier, J.; Norquist, A. J. Nature 2016, 533, 73. doi: 10.1038/nature17439
11. [11]
  (11) Roszak, R.; Beker, W.; Molga, K.; Grzybowski, B. A. J. Am. Chem. Soc. 2019, 141, 17142. doi: 10.1021/jacs.9b05895(11) Roszak, R.; Beker, W.; Molga, K.; Grzybowski, B. A. J. Am. Chem. Soc. 2019, 141, 17142. doi: 10.1021/jacs.9b05895
12. [12]
  (12) Gao, H.; Struble, T. J.; Coley, C. W.; Wang, Y.; Green, W. H.; Jensen, K. F. ACS Central Sci. 2018, 4, 1465. doi: 10.1021/acscentsci.8b00357(12) Gao, H.; Struble, T. J.; Coley, C. W.; Wang, Y.; Green, W. H.; Jensen, K. F. ACS Central Sci. 2018, 4, 1465. doi: 10.1021/acscentsci.8b00357
13. [13]
  (13) Zahrt, A. F.; Henle, J. J.; Rose, B. T.; Wang, Y.; Darrow, W. T.; Denmark, S. E. Science 2019, 363, 1. doi: 10.1126/science.aau5631(13) Zahrt, A. F.; Henle, J. J.; Rose, B. T.; Wang, Y.; Darrow, W. T.; Denmark, S. E. Science 2019, 363, 1. doi: 10.1126/science.aau5631
14. [14]
  (14) Reid, J. P.; Sigman, M. S. Nature 2019, 571, 343. doi: 10.1038/s41586-019-1384-z(14) Reid, J. P.; Sigman, M. S. Nature 2019, 571, 343. doi: 10.1038/s41586-019-1384-z
15. [15]
  (15) Segler, M. H. S.; Preuss, M.; Waller, M. P. Nature 2018, 555, 604. doi: 10.1038/nature25978(15) Segler, M. H. S.; Preuss, M.; Waller, M. P. Nature 2018, 555, 604. doi: 10.1038/nature25978
16. [16]
  (16) Coley, C. W.; Thomas, D. A.; Lummiss, J. A. M.; Jaworski, J. N.; Breen, C. P.; Schultz, V.; Hart, T.; Fishman, J. S.; Rogers, L,; Gao, H.; et al. Science 2019, 365, 1. doi: 10.1126/science.aax1566(16) Coley, C. W.; Thomas, D. A.; Lummiss, J. A. M.; Jaworski, J. N.; Breen, C. P.; Schultz, V.; Hart, T.; Fishman, J. S.; Rogers, L,; Gao, H.; et al. Science 2019, 365, 1. doi: 10.1126/science.aax1566
17. [17]
  (17) Santanilla, A. B.; Regalado, E. L.; Pereira, T.; Shevlin, M.; Bateman, K.; Campeau, L. C.; Schneeweis, J.; Berritt, S.; Shi, Z. C.; Nantermet, P.; et al. Science 2015, 347, 49. doi: 10.1126/science.1259203(17) Santanilla, A. B.; Regalado, E. L.; Pereira, T.; Shevlin, M.; Bateman, K.; Campeau, L. C.; Schneeweis, J.; Berritt, S.; Shi, Z. C.; Nantermet, P.; et al. Science 2015, 347, 49. doi: 10.1126/science.1259203
18. [18]
  (18) Krska, S. W.; DiRocco, D. A.; Dreher, S. D.; Shevlin, M. Accounts Chem. Res. 2017, 50, 2976. doi: 10.1021/acs.accounts.7b00428(18) Krska, S. W.; DiRocco, D. A.; Dreher, S. D.; Shevlin, M. Accounts Chem. Res. 2017, 50, 2976. doi: 10.1021/acs.accounts.7b00428
19. [19]
  (19) Mennen, S. M.; Alhambra, C.; Allen, C. L.; Barberis, M.; Berritt, S.; Brandt, T. A.; Campbell, A. D.; Castañón, J.; Cherney, A. H.; Christensen, M.; et al. Org. Process Res. Dev. 2019, 23, 1213. doi: 10.1021/acs.oprd.9b00140(19) Mennen, S. M.; Alhambra, C.; Allen, C. L.; Barberis, M.; Berritt, S.; Brandt, T. A.; Campbell, A. D.; Castañón, J.; Cherney, A. H.; Christensen, M.; et al. Org. Process Res. Dev. 2019, 23, 1213. doi: 10.1021/acs.oprd.9b00140
20. [20]
  (20) Seefried, F.; Schmidt, T.; Reinecke, M.; Heinzlmeir, S.; Kuster, B.; Wilhelm, M. J. Proteome Res. 2019, 18, 1486. doi: 10.1021/acs.jproteome.8b00724(20) Seefried, F.; Schmidt, T.; Reinecke, M.; Heinzlmeir, S.; Kuster, B.; Wilhelm, M. J. Proteome Res. 2019, 18, 1486. doi: 10.1021/acs.jproteome.8b00724
21. [21]
  (21) Figueiredo, R. M.; Suppo, J. S.; Campagne, J. M. Chem. Rev. 2016, 116, 12029. doi: 10.1021/acs.chemrev.6b00237(21) Figueiredo, R. M.; Suppo, J. S.; Campagne, J. M. Chem. Rev. 2016, 116, 12029. doi: 10.1021/acs.chemrev.6b00237
22. [22]
  (22) Roughley, S. D.; Jordan, A. M. J. Med. Chem. 2011, 54, 3451. doi: 10.1021/jm200187y(22) Roughley, S. D.; Jordan, A. M. J. Med. Chem. 2011, 54, 3451. doi: 10.1021/jm200187y
23. [23]
  (23) Sabatini, M. T.; Boulton, L. T.; Sneddon, H. F.; Sheppard, T. D. Nat. Catal. 2019, 2, 10. doi: 10.1038/s41929-018-0211-5(23) Sabatini, M. T.; Boulton, L. T.; Sneddon, H. F.; Sheppard, T. D. Nat. Catal. 2019, 2, 10. doi: 10.1038/s41929-018-0211-5
24. [24]
  (24) Brown, D. G.; Bostrom, J. J. Med. Chem. 2016, 59, 4443. doi: 10.1021/acs.jmedchem.5b01409(24) Brown, D. G.; Bostrom, J. J. Med. Chem. 2016, 59, 4443. doi: 10.1021/acs.jmedchem.5b01409
25. [25]
  (25) Halford, B. ACS Central Sci. 2022, 8, 405. doi: 10.1021/acscentsci.2c00369(25) Halford, B. ACS Central Sci. 2022, 8, 405. doi: 10.1021/acscentsci.2c00369
26. [26]
  (26) Syed, Y. Y. Drugs 2022, 82, 455. doi: 10.1007/s40265-022-01684-5(26) Syed, Y. Y. Drugs 2022, 82, 455. doi: 10.1007/s40265-022-01684-5
27. [27]
  (27) Ghosh, S. C.; Ngiam, J. S.; Seayad, A. M.; Tuan, D. T.; Chai, C. L. L.; Chen, A. J. Org. Chem. 2012, 77, 8007. doi: 10.1021/jo301252c(27) Ghosh, S. C.; Ngiam, J. S.; Seayad, A. M.; Tuan, D. T.; Chai, C. L. L.; Chen, A. J. Org. Chem. 2012, 77, 8007. doi: 10.1021/jo301252c
28. [28]
  (28) Pattabiraman, V. R.; Bode, J. W. Nature 2011, 480, 471. doi: 10.1038/nature10702(28) Pattabiraman, V. R.; Bode, J. W. Nature 2011, 480, 471. doi: 10.1038/nature10702
29. [29]
  (29) Beker, W.; Gajewska, E. P.; Badowski, T.; Grzybowski, B. A. Angew. Chem.-Int. Edit. 2019, 58, 4515. doi: 10.1002/anie.201806920(29) Beker, W.; Gajewska, E. P.; Badowski, T.; Grzybowski, B. A. Angew. Chem.-Int. Edit. 2019, 58, 4515. doi: 10.1002/anie.201806920
30. [30]
  (30) Aydogdu, S.; Hatipoglu, A. J. Indian Chem. Soc. 2022, 99, 100752. doi: 10.1016/j.jics.2022.100752(30) Aydogdu, S.; Hatipoglu, A. J. Indian Chem. Soc. 2022, 99, 100752. doi: 10.1016/j.jics.2022.100752
31. [31]
  (31) Ma, Y.; Zhang, X.; Zhu, L.; Feng, X.; Kowah, J. A. H.; Jiang, J.; Wang, L.; Jiang, L.; Liu, X. Molecules 2023, 28, 5995. doi: 10.3390/molecules28165995(31) Ma, Y.; Zhang, X.; Zhu, L.; Feng, X.; Kowah, J. A. H.; Jiang, J.; Wang, L.; Jiang, L.; Liu, X. Molecules 2023, 28, 5995. doi: 10.3390/molecules28165995
32. [32]
  (32) Ramakrishnan, R.; Dral, P. O.; Rupp, M.; Lilienfeld, O. A. V. Sci. Data 2014, 1, 140022. doi: 10.1038/sdata.2014.22(32) Ramakrishnan, R.; Dral, P. O.; Rupp, M.; Lilienfeld, O. A. V. Sci. Data 2014, 1, 140022. doi: 10.1038/sdata.2014.22
33. [33]
  (33) Tsubaki, M.; Mizoguchi, T. J. Phys. Chem. Lett. 2018, 9, 5733. doi: 10.1021/acs.jpclett.8b01837(33) Tsubaki, M.; Mizoguchi, T. J. Phys. Chem. Lett. 2018, 9, 5733. doi: 10.1021/acs.jpclett.8b01837
34. [34]
  (34) https://github.com/doylelab/rxnpredict (accessed Dec. 28, 2023)(34) https://github.com/doylelab/rxnpredict (accessed Dec. 28, 2023)
35. [35]
  (35) Yousef, W. A. Pattern Recognit. Lett. 2021, 146, 115. doi: 10.1016/j.patrec.2021.02.022(35) Yousef, W. A. Pattern Recognit. Lett. 2021, 146, 115. doi: 10.1016/j.patrec.2021.02.022
36. [36]
  (36) Dodge, Y. The Concise Encyclopedia of Statistics; Springer New York: New York, NY, USA, 2008; pp. 88–91.(36) Dodge, Y. The Concise Encyclopedia of Statistics; Springer New York: New York, NY, USA, 2008; pp. 88–91.
37. [37]
  (37) Zollanvari, A.; Dougherty, E. R. Pattern Recognit. 2014, 47, 2178. doi: 10.1016/j.patcog.2013.11.022(37) Zollanvari, A.; Dougherty, E. R. Pattern Recognit. 2014, 47, 2178. doi: 10.1016/j.patcog.2013.11.022
38. [38]
  (38) Song, W.; Dong, K.; Li, M. Org. Lett. 2020, 22, 371. doi: 10.1021/acs.orglett.9b03905(38) Song, W.; Dong, K.; Li, M. Org. Lett. 2020, 22, 371. doi: 10.1021/acs.orglett.9b03905
39. [39]
  (39) Mali, S. M.; Bhaisare, R. D.; Gopi, H. N. J. Org. Chem. 2013, 78, 5550. doi: 10.1021/jo400701v(39) Mali, S. M.; Bhaisare, R. D.; Gopi, H. N. J. Org. Chem. 2013, 78, 5550. doi: 10.1021/jo400701v
40. [40]
  (40) Chen, Z.; Fu, R.; Chai, W.; Zheng, H.; Sun, L.; Lu, Q.; Yuan, R. Tetrahedron 2014, 70, 2237. doi: 10.1016/j.tet.2014.02.042(40) Chen, Z.; Fu, R.; Chai, W.; Zheng, H.; Sun, L.; Lu, Q.; Yuan, R. Tetrahedron 2014, 70, 2237. doi: 10.1016/j.tet.2014.02.042
41. [41]
  (41) Li, X.; Li, Z.; Deng, H.; Deng, H.; Zhou, X. Tetrahedron Lett. 2013, 54, 2212. doi: 10.1016/j.tetlet.2013.02.058(41) Li, X.; Li, Z.; Deng, H.; Deng, H.; Zhou, X. Tetrahedron Lett. 2013, 54, 2212. doi: 10.1016/j.tetlet.2013.02.058

扫一扫看文章

计量

PDF下载量: 3
文章访问数: 89
HTML全文浏览量: 14

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

基于小数据集的机器学习预测酰胺键合成转化率

通讯作者: 张利静, E-mail: zhanglj@dlut.edu.cn; 陶胜洋, E-mail: taosy@dlut.edu.cn

English

Machine Learning Enables the Prediction of Amide Bond Synthesis Based on Small Datasets

Corresponding author: Lijing Zhang, ; Shengyang Tao,

CCS Chemistry：嵌段共聚物自组装成 “三明治”二维有机材料，实现纳米级压力传感功能

CCS Chemistry：颜徐州、鲍哲南：你的皮肤可能是一片SPMs

CCS Chemistry：工作高效不疲劳，只须让催化剂动起来

CCS Chemistry：静电组装导向聚合- 聚电解质纳米凝胶量化制备新策略

CCS Chemistry：金黄色葡萄球菌醛基脱氢酶是如何识别特异性底物的？

计量

通讯作者: 陈斌, bchen63@163.com

中国化学会秘书处

基于小数据集的机器学习预测酰胺键合成转化率

通讯作者: 张利静, E-mail: zhanglj@dlut.edu.cn; 陶胜洋, E-mail: taosy@dlut.edu.cn

English

Machine Learning Enables the Prediction of Amide Bond Synthesis Based on Small Datasets

Corresponding author: Lijing Zhang, ; Shengyang Tao,

CCS Chemistry：嵌段共聚物自组装成 “三明治”二维有机材料，实现纳米级压力传感功能

CCS Chemistry：颜徐州、鲍哲南： 你的皮肤可能是一片SPMs

CCS Chemistry：工作高效不疲劳，只须让催化剂动起来

CCS Chemistry：静电组装导向聚合- 聚电解质纳米凝胶量化制备新策略

CCS Chemistry：金黄色葡萄球菌醛基脱氢酶是如何识别特异性底物的？

计量

文章相关

通讯作者: 陈斌, bchen63@163.com

中国化学会秘书处

CCS Chemistry：颜徐州、鲍哲南：你的皮肤可能是一片SPMs