Prediction of pharmacological activities from chemical structures with graph convolutional neural networks

1. Krizhevsky, A., Sutskever, I. & Hinton, G. Imagenet classification with deep convolutional neural networks. Adv. Neural Inf. Process.

Syst. 60, 1097–1105 (2012).

2. Duvenaud, D. et al. Convolutional networks on graphs for learning molecular fingerprints. Adv. Neural Inf. Process. Syst. 2015,

2224–2232 (2015).

3. Wu, Z. et al. MoleculeNet: A benchmark for molecular machine learning. Chem. Sci. 9, 513–530 (2018).

4. DeepChem. https​://githu​b.com/deepc​hem/deepc​hem. Accessed 21 Apr 2019.

5. Altae-Tran, H., Ramsundar, B., Pappu, A. S. & Pande, V. Low data drug discovery with one-shot learning. ACS Cent. Sci. 3, 283–293

(2017).

6. Cai, C. et al. Deep learning-based prediction of drug-induced cardiotoxicity. J. Chem. Inf. Model. 59, 1073–1084 (2019).

7. Cheng, W. & Ng, C. A. Using machine learning to classify bioactivity for 3486 per- and polyfluoroalkyl substances (PFASs) from

the OECD list. Environ. Sci. Technol. 53, 13970–13980 (2019).

8. Rodríguez-Pérez, R., Miyao, T., Jasial, S., Vogt, M. & Bajorath, J. Prediction of compound profiling matrices using machine learning. ACS Omega 3, 4713–4723 (2018).

9. Miyazaki, Y., Ono, N., Huang, M., Altaf-Ul-Amin, M. & Kanaya, S. Comprehensive exploration of target-specific ligands using a

graph convolution neural network. Mol. Inf. 39, 1900095 (2020).

10. Mayr, A. et al. Large-scale comparison of machine learning methods for drug target prediction on ChEMBL. Chem. Sci. 9,

5441–5451 (2018).

11. Bosc, N. et al. Large scale comparison of QSAR and conformal prediction methods and their applications in drug discovery. J.

Cheminform. 11, 4 (2019).

12. Unterthiner, T. et al. Deep learning as an opportunity in virtual screening. Adv. Neural Inf. Process. Syst. 27, 1–9 (2014).

13. Gomes, J., Ramsundar, B., Feinberg, E. N. & Pande, V. S. Atomic convolutional networks for predicting protein-ligand binding

affinity. Preprint at https​://arxiv​.org/abs/1703.10603​ (2017).

14. Karimi, M., Wu, D., Wang, Z. & Shen, Y. DeepAffinity: interpretable deep learning of compound–protein affinity through unified

recurrent and convolutional neural networks. Bioinformatics 35, 3329–3338 (2019).

15. Öztürk, H., Özgür, A. & Ozkirimli, E. DeepDTA: Deep drug-target binding affinity prediction. Bioinformatics 34, i821–i829 (2018).

16. Wang, X. et al. Dipeptide frequency of word frequency and graph convolutional networks for DTA prediction. Front. Bioeng.

Biotechnol. 8, 267 (2020).

17. Liu, P., Li, H., Li, S. & Leung, K.-S. Improving prediction of phenotypic drug response on cancer cell lines using deep convolutional

network. BMC Bioinform. 20, 1–14 (2019).

18. Whitehead, T. M., Irwin, B. W. J., Hunt, P., Segall, M. D. & Conduit, G. J. Imputation of assay bioactivity data using deep learning.

J. Chem. Inf. Model. 59, 1197–1204 (2019).

19. Feinberg, E. N. et al. PotentialNet for molecular property prediction. ACS Cent. Sci. 4, 1520–1530 (2018).

20. Karlov, D. S., Sosnin, S., Fedorov, M. V. & Popov, P. GraphDelta: MPNN scoring function for the affinity prediction of proteinligand complexes. ACS Omega 5, 5150–5159 (2020).

21. Wu, J. et al. Precise modelling and interpretation of bioactivities of ligands targeting G protein-coupled receptors. Bioinformatics

35, i324–i332 (2019).

22. Wang, X. et al. Molecule property prediction based on spatial graph embedding. J. Chem. Inf. Model. 59, 3817–3828 (2019).

23. Lenselink, E. B. et al. Beyond the hype: Deep neural networks outperform established methods using a ChEMBL bioactivity

benchmark set. J. Cheminform. 9, 45 (2017).

24. ChemAxon. https​://chema​xon.com. Accessed 11 Jan 2019.

25. RDKit, Open-Source Chemiformatics Software. http://www.rdkit​.org. Accessed 21 Apr 2019.

26. Jiménez, J. & Ginebra, J. pyGPGO: Bayesian optimization for python. J. Open Source Softw. 2, 431 (2017).

Scientific Reports |

Vol:.(1234567890)

(2021) 11:525 |

https://doi.org/10.1038/s41598-020-80113-7

12

www.nature.com/scientificreports/

A Self-archived copy in

Kyoto University Research Information Repository

https://repository.kulib.kyoto-u.ac.jp

27. Xu, Y., Pei, J. & Lai, L. Deep learning based regression and multiclass models for acute oral toxicity prediction with automatic

chemical feature extraction. J. Chem. Inf. Model. 57, 2672–2685 (2017).

28. Yang, K. et al. Analyzing learned molecular representations for property prediction. J. Chem. Inf. Model. 59, 3370–3388 (2019).

29. Kwon, S., Bae, H., Jo, J. & Yoon, S. Comprehensive ensemble in QSAR prediction for drug discovery. BMC Bioinform. 20, 1–12

(2019).

30. Bemis, G. W. & Murcko, M. A. The properties of known drugs. 1. Molecular frameworks. J. Med. Chem. 39, 2887–2893 (1996).

31. Godden, J. W. & Bajorath, J. Differential Shannon entropy as a sensitive measure of differences in database variability of molecular

descriptors. J. Chem. Inf. Comput. Sci. 41, 1060–1066 (2001).

32. Schneider, P. & Schneider, G. Privileged structures revisited. Angew. Chem. Int. Ed. 56, 7971–7974 (2017).

33. Asano, M. et al. SKF-10047, a prototype Sigma-1 receptor agonist, augmented the membrane trafficking and uptake activity of

the serotonin transporter and its C-terminus-deleted mutant via a Sigma-1 receptor-independent mechanism. J. Pharmacol. Sci.

139, 29–36 (2019).

34. Ramamoorthy, S. et al. Antidepressant- and cocaine-sensitive human serotonin transporter: Molecular cloning, expression, and

chromosomal localization. Proc. Natl. Acad. Sci. U.S.A. 90, 2542–2546 (1993).

35. Nishitani, N. et al. Manipulation of dorsal raphe serotonergic neurons modulates active coping to inescapable stress and anxietyrelated behaviors in mice and rats. Neuropsychopharmacology 44, 721–732 (2019).

36. Mervin, L. H. et al. Target prediction utilising negative bioactivity data covering large chemical space. J. Cheminform. 7, 1–16

(2015).

37. Romeo, G. et al. New pyrimido[5,4-b]indoles as ligands for α1-adrenoceptor subtypes. J. Med. Chem. 46, 2877–2894 (2003).

38. Koutsoukas, A., Monaghan, K. J., Li, X. & Huan, J. Deep-learning: Investigating deep neural networks hyper-parameters and

comparison of performance to shallow methods for modeling bioactivity data. J. Cheminform. 9, 1–13 (2017).

39. Willmott, C. & Matsuura, K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing

average model performance. Clim. Res. 30, 79–82 (2005).

40. Chai, T. & Draxler, R. R. Root mean square error (RMSE) or mean absolute error (MAE)? Arguments against avoiding RMSE in

the literature. Geosci. Model Dev. 7, 1247–1250 (2014).

41. Li, Q., Han, Z. & Wu, X. M. Deeper insights into graph convolutional networks for semi-supervised learning. In AAAI 2018 (2018).

42. Goh, G. B., Siegel, C., Vishnu, A., Hodas, N. O. & Baker, N. Chemception: A deep neural network with minimal chemistry knowledge matches the performance of expert-developed QSAR/QSPR models. Preprint at https​://arxiv​.org/abs/1706.06689​ (2017)

43. Cortés-Ciriano, I. & Bender, A. KekuleScope: Prediction of cancer cell line sensitivity and compound potency using convolutional

neural networks trained on compound images. J. Cheminform. 11, 41 (2019).

44. Uesawa, Y. Quantitative structure–activity relationship analysis using deep learning based on a novel molecular image input

technique. Bioorg. Med. Chem. Lett. 28, 3400–3403 (2018).

45. Hirohara, M., Saito, Y., Koda, Y., Sato, K. & Sakakibara, Y. Convolutional neural network based on SMILES representation of

compounds for detecting chemical motif. BMC Bioinform. https​://doi.org/10.1186/s1285​9-018-2523-5 (2018).

46. Nidhi, G. M., Davies, J. W. & Jenkins, J. L. Prediction of biological targets for compounds using multiple-category Bayesian models

trained on chemogenomics databases. J. Chem. Inf. Model. 46, 1124–1133 (2006).

47. Shang, J. et al. Comparative analyses of structural features and scaffold diversity for purchasable compound libraries. J. Cheminform.

9, 25 (2017).

48. Li, Y., Zhang, L. & Liu, Z. Multi-objective de novo drug design with conditional graph generative model. J. Cheminform. 10, 33

(2018).

49. Paricharak, S. et al. Data-driven approaches used for compound library design, hit triage and bioactivity modeling in highthroughput screening. Brief. Bioinform. 19, 277–285 (2018).

50. Zhang, Y. & Lee, A. A. Bayesian semi-supervised learning for uncertainty-calibrated prediction of molecular properties and active

learning. Chem. Sci. 10, 8154–8163 (2019).

51. Robinson, M. C., Glen, R. C. & Lee, A. A. Validating the validation: reanalyzing a large-scale comparison of deep learning and

machine learning models for bioactivity prediction. J. Comput. Aided. Mol. Des. 34, 717–730 (2020).

52. Tatsumi, M., Groshan, K., Blakely, R. D. & Richelson, E. Pharmacological profile of antidepressants and related compounds at

human monoamine transporters. Eur. J. Pharmacol. 340, 249–258 (1997).

53. Tarasova, O. A. et al. QSAR modeling using large-scale databases: Case study for HIV-1 reverse transcriptase inhibitors. J. Chem.

Inf. Model. 55, 1388–1399 (2015).

Acknowledgements

We thank Dr. Randy Blakely (Florida Atlantic University) for providing hSERT-pcDNA3. We also thank the

ChemAxon for the free academic license of Instant J Chem. This work was partly supported by Grants-in-Aid

for Scientific Research from JSPS (to K.N. (JP20H04774, JP20K07064), to S.K. (JP18H04616, JP20H00491)),

AMED (to S.K. (JP20ak0101088h0003)), and SENSHIN Medical Research Foundation (to K.N.).

Author contributions

M.S. wrote the code, performed the experiments, and wrote the main manuscript. K.N. initiated the experiments and wrote the manuscript for assay related part. N.S., C.A., K.T., and H.S. conducted the in vitro and

behavioral tests. S.K. and K.N. obtained funding. K.N. and S.K. supervised the study. M.S., K.N., and S.K. edited

the manuscript.

Competing interests M.S. is an employee of Medical Database Ltd. The other authors declare no competing interests.

Additional information

Supplementary Information The online version contains supplementary material available at https​://doi.

org/10.1038/s4159​8-020-80113​-7.

Correspondence and requests for materials should be addressed to K.N. or S.K.

Reprints and permissions information is available at www.nature.com/reprints.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and

institutional affiliations.

Scientific Reports |

(2021) 11:525 |

https://doi.org/10.1038/s41598-020-80113-7

13

Vol.:(0123456789)

www.nature.com/scientificreports/

A Self-archived copy in

Kyoto University Research Information Repository

https://repository.kulib.kyoto-u.ac.jp

Open Access This article is licensed under a Creative Commons Attribution 4.0 International

License, which permits use, sharing, adaptation, distribution and reproduction in any medium or

format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the

Creative Commons licence, and indicate if changes were made. The images or other third party material in this

article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the

material. If material is not included in the article’s Creative Commons licence and your intended use is not

permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from

the copyright holder. To view a copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/.

© The Author(s) 2021

Scientific Reports |

Vol:.(1234567890)

(2021) 11:525 |

https://doi.org/10.1038/s41598-020-80113-7

14

...