Abstract
Protein-protein interactions play an important role in the development of new therapeutic treatments and prophylactic vaccines. For instance, the efficacy of a vaccine strongly depends to what extent an antibody may form a stable bond with an antigen. In-laboratory experiments are both time-consuming and expensive, which limits their scope to only the most relevant interactions. Computational experiments, on the other hand, have the potential to explore and screen a vast number of possibilities, thus providing experimentalists with the most promising cases. Protein-protein interactions may be learned by deep learning networks. Nonetheless, the training of these networks requires a large number of instances which may not be readily available, as in the case, of rare or new diseases. Furthermore, the learning process is made more complex by the scarcity of data about non-interacting proteins, making the network prone to overfitting. These two shortcomings are addressed in this paper. A new learnable pyramid network architecture is proposed in which the depth and the complexity of the network are directly learned from the data, which makes the network suitable for both small and large datasets. Our network outperforms state-of-the-art neural networks for protein-protein interaction predictions and is capable of adapting its architecture to datasets whose size varies from a few thousand to seventeen million. A classification accuracy of over 96% is achieved for all datasets.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Alberts, B.: The cell as a collection of protein machines: preparing the next generation of molecular biologists. Cell 92(3), 291–294 (1998)
Bejani, M.M., Ghatee, M.: A systematic review on overfitting control in shallow and deep neural networks. Artif. Intell. Rev. 54(8), 6391–6438 (2021). https://doi.org/10.1007/s10462-021-09975-1
Chen, M., Ju, C., Zhou, G., Chen, X., Zhang, T., Chang, K., Zaniolo, C., Wang, W.: Multifaceted protein-protein interaction prediction based on siamese residual RCNN. Bioinformatics 35(14), i305–i314 (2019)
Chin, M., Marks, C., Deane, C.M.: Humanization of antibodies using a machine learning approach on large-scale repertoire data. bioRxiv (2021)
Deng, J., Dong, W., Socher, R., Li, L., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 248–255. Miami, FL, USA (2009)
Fukushima, K., Miyake, S.: Neocognitron: a self-organizing neural network model for a mechanism of visual pattern recognition. In: Competition and Cooperation in Neural Nets, pp. 267–285. Berlin, Heidelberg (1982)
Guo, Y., et al.: PRED_PPI: a server for predicting protein-protein interactions based on sequence data with probability assignment. BMC. Res. Notes 3(1), 1–7 (2010)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778. Las Vegas, NV, USA (2016)
Ioffe, S., Szegedy, C.: Batch Normalization: accelerating deep network training by reducing internal covariate shift. In: International Conference on Machine Learning, vol. 37, pp. 448–456. Lille, France (2015)
Li, H., Gong, X.J., Yu, H., Zhou, C.: Deep neural network based predictions of protein interactions using primary sequences. Molecules 23(8), 1923–1939 (2018)
Li, H., Xu, Z., Taylor, G., Studer, C., Goldstein, T.: Visualizing the loss landscape of neural nets. arXiv preprint arXiv:1712.09913 (2017)
Li, T., et al.: A scored human protein-protein interaction network to catalyze genomic interpretation. Nat. Methods 14(1), 61–78 (2017)
Liang, S., Zhang, C.: Prediction of immunogenicity for humanized and full human therapeutic antibodies. PLoS ONE 15, 1–14 (2020)
Meyer, Y.: Wavelets and Operators, vol. 1. Cambridge University Press, Cambridge (1992)
Neek, M., Kim, T.I., Wang, S.W.: Protein-based nanoparticles in cancer vaccine development. Nanomed. Nanotechnol. Biol. Med. 15(1), 164–174 (2019)
Neiswinger, J., et al.: Protein microarrays: flexible tools for scientific innovation. Cold Spring Harbor Protocols 2016(10), pdb-top081471 (2016)
Orhan, E., Pitkow, X.: Skip connections eliminate singularities. In: International Conference on Learning Representations. Vancouver, Canada (2018)
Schaefer, M.H., Fontaine, J.F., Vinayagam, A., Porras, P., Wanker, E.E., Andrade-Navarro, M.A.: HIPPIE: integrating protein interaction networks with experiment based quality scores. PLoS ONE 7(2), 1–8 (2012)
Sun, T., Zhou, B., Lai, L., Pei, J.: Sequence-based prediction of protein protein interaction using a deep-learning algorithm. BMC Bioinf. 18(1), 1–8 (2017)
Szklarczyk, D., et al.: STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 47(D1), D607–D613 (2019)
Titeca, K., Lemmens, I., Tavernier, J., Eyckerman, S.: Discovering cellular protein-protein interactions: technological strategies and opportunities. Mass spectrometry Rev. 38(1), 79–111 (2019)
Xenarios, I., Salwinski, L., Duan, X.J., Higney, P., Kim, S.M., Eisenberg, D.: DIP, the database of interacting proteins: a research tool for studying cellular networks of protein interactions. Nucleic Acids Res. 30(1), 303–305 (2002)
Zeng, M., Zhang, F., Wu, F., Li, Y., Wang, J., Li, M.: Protein-protein interaction site prediction through combining local and global features with deep neural networks. Bioinformatics 36(4), 1114–1120 (2020)
Zhang, C., Bengio, S., Hardt, M., Recht, B., Vinyals, O.: Understanding deep learning (still) requires rethinking generalization. Commun. ACM 64(3), 107–115 (2021)
Zhu, H., et al.: Global analysis of protein activities using proteome chips. Science 293(5537), 2101–2105 (2001)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Wu, J., Paquet, E., Viktor, H.L., Michalowski, W. (2023). Adapting to Complexity: Deep Learnable Architecture for Protein-protein Interaction Predictions. In: Nicosia, G., et al. Machine Learning, Optimization, and Data Science. LOD 2022. Lecture Notes in Computer Science, vol 13810. Springer, Cham. https://doi.org/10.1007/978-3-031-25599-1_39
Download citation
DOI: https://doi.org/10.1007/978-3-031-25599-1_39
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-25598-4
Online ISBN: 978-3-031-25599-1
eBook Packages: Computer ScienceComputer Science (R0)