Abstract
Not only is modeling in-vivo protein-DNA binding basic to a deeper comprehension of regulatory mechanisms, but a complicated job in computational biology. Although current deep-learning based methods have achieved some success in-vivo protein-DNA binding, on the one hand, they tend to ignore the weakly supervised information genome sequences, that is, the bound DNA sequence has a high probability of containing more than one TFBS. On the other hand, One-hot encoding requires each category to be independent of each other, and the dependence between nucleotides is ignored when it is used to encode DNA sequences. In order to solve this problem, we developed a framework based on weakly-supervised. The structure proposed in this paper combines multi-instance learning with hybrid deep neural networks and uses K-mer encoding instead of one-hot encoding to process DNA sequences, this operation simulates in-vivo protein-DNA binding. First of all, we use the concepts of MIL to segments the input sequence into many overlapping instances, and then use K-mer encoding to convert these instances into high-order dependent inputs of the image-like. Then hybrid deep neural network that integrates convolutional and recurrent neural networks is used to calculate the score of all the instances contained in the same bag. Finally, it uses the “Noisy-and” method to integrate the predicted values for all instances into the final predicted values for the bag. This paper discusses the effect of K-mer encoding on the function of the framework and verifies the function of “Noisy-and” compared with other fusion methods.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Elnitski, L., Jin, V.X., Farnham, P.J., Jones, S.J.M.: Locating mammalian transcription factor binding sites: a survey of computational and experimental techniques. Genome Res. 16, 1455–1464 (2006)
Orenstein, Y., Shamir, R.: A comparative analysis of transcription factor binding models learned from PBM, HT-SELEX and ChIP data. Nucleic Acids Res. 42, e63–e63 (2014)
Furey, T.S.: ChIP–seq and beyond: new and improved methodologies to detect and characterize protein–DNA interactions. Nat. Rev. Genet. 13, 840–852 (2012)
Jothi, R., Cuddapah, S., Barski, A., Cui, K., Zhao, K.: Genome-wide identification of in vivo protein–DNA binding sites from ChIP-Seq data. Nucleic Acids Res. 36, 5221–5231 (2008)
Stormo, G.D.: Consensus patterns in DNA. Methods Enzymol. 183, 211–221 (1990)
Stormo, G.D.: DNA binding sites: representation and discovery. Bioinformatics 16, 16–23 (2000)
Zhao, X., Huang, H., Speed, T.P.: Finding short DNA motifs using permuted Markov models. J. Comput. Biol. 12, 894–906 (2005)
Badis, G., et al.: Diversity and complexity in DNA recognition by transcription factors. Science 324, 1720–1723 (2009)
Ghandi, M., et al.: gkmSVM: an R package for gapped-kmer SVM. Bioinformatics 32, 2205–2207 (2016)
Alipanahi, B., Delong, A., Weirauch, M.T., Frey, B.J.: Predicting the sequence specificities of DNA-and RNA-binding proteins by deep learning. Nat. Biotechnol. 33, 831–838 (2015)
Zhou, J., Troyanskaya, O.G.: Predicting effects of noncoding variants with deep learning-based sequence model. Nat. Methods 12, 931–934 (2015)
Quang, D., Xie, X.: DanQ: a hybrid convolutional and recurrent deep neural network for quantifying the function of DNA sequences. Nucleic Acids Res. 44, e107–e107 (2016)
Zeng, H., Edwards, M.D., Liu, G., Gifford, D.K.: Convolutional neural network architectures for predicting DNA–protein binding. Bioinformatics 32, i121–i127 (2016)
Kelley, D.R., Snoek, J., Rinn, J.L.: Basset: learning the regulatory code of the accessible genome with deep convolutional neural networks. Genome Res. 26, 990–999 (2016)
Hassanzadeh, H.R., Wang, M.D.: DeeperBind: enhancing prediction of sequence specificities of DNA binding proteins. In: IEEE International Conference on Bioinformatics and Biomedicine, pp. 178–183 (2017)
Shrikumar, A., Greenside, P., Kundaje, A.: Reverse-complement parameter sharing improves deep learning models for genomics. bioRxiv, 103663 (2017)
Lo Bosco, G., Di Gangi, M.: Deep learning architectures for DNA sequence classification. In: Petrosino, A., Loia, V., Pedrycz, W. (eds.) WILF 2016. LNCS (LNAI), vol. 10147, pp. 162–171. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-52962-2_14
Gao, Z., Ruan, J.: Computational modeling of in vivo and in vitro protein-DNA interactions by multiple instance learning. Bioinformatics 33(14), 2097–2105 (2017)
Annala, M., Laurila, K., Lähdesmäki, H., Nykter, M.: A linear model for transcription factor binding affinity prediction in protein binding microarrays. PloS One 6, e20059 (2011)
Zhang, Q., Zhu, L., Bao, W., Huang, D.S.: Weakly supervised convolutional neural network architecture for predicting protein-DNA binding. IEEE/ACM Trans. Comput. Biol. Bioinform. 17, 679–689 (2018)
Keilwagen, J., Grau, J.: Varying levels of complexity in transcription factor binding motifs. Nucleic Acids Res. 43, e119 (2015)
Siebert, M., Söding, J.: Bayesian Markov models consistently outperform PWMs at predicting motifs in nucleotide sequences. Nucleic Acids Res. 44, 6055–6069 (2016)
Eggeling, R., Roos, T., Myllymäki, P., Grosse, I.: Inferring intra-motif dependencies of DNA binding sites from ChIP-seq data. BMC Bioinformatics 16, 1–15 (2015)
Zhou, T., et al.: Quantitative modeling of transcription factor binding specificities using DNA shape. Proc. Natl. Acad. Sci. 112(15), 4654–4659 (2015)
Zhang, Q., Zhu, L., Huang, D.S.: High-order convolutional neural network architecture for predicting DNA-protein binding sites. IEEE/ACM Trans. Comput. Biol. Bioinf. 1, 1–1 (2018)
Kraus, O.Z., Ba, J.L., Frey, B.J.: Classifying and segmenting microscopy images with deep multiple instance learning. Bioinformatics 32, i52–i59 (2016)
Huang, D.S.: Systematic Theory of Neural Networks for Pattern Recognition, vol. 201. Publishing House of Electronic Industry of China, Beijing (1996)
Huang, D.S.: Radial basis probabilistic neural networks: model and application. Int. J. Pattern Recogn. Artif. Intell. 13, 1083–1101 (1999)
Huang, D.S., Du, J.X.: A constructive hybrid structure optimization methodology for radial basis probabilistic neural networks. IEEE Trans. Neural Netw. 19, 2099–2115 (2008)
Glorot, X., Bordes, A., Bengio, Y.: Deep sparse rectifier neural networks. In: Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, pp. 315–323 (2011)
Srivastava, N., Hinton, G.E., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2014)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9, 1735–1780 (1997)
Durand, T., Thome, N., Cord, M.: WELDON: weakly supervised learning of deep convolutional neural networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4743–4752 (2016)
Fawcett, T.: An introduction to ROC analysis. Pattern Recogn. Lett. 27, 861–874 (2006)
Davis, J., Goadrich, M.: The relationship between precision-recall and ROC curves. In: ICML 2006: Proceedings of the International Conference on Machine Learning, New York, NY, USA, pp. 233–240 (2006)
Sasaki, Y.: The truth of the F-measure. Teach. Tutor. Mater. 1(5), 1–5 (2007)
Glorot, X., Bengio, Y.: Understanding the difficulty of training deep feedforward neural networks. J. Mach. Learn. Res. 9, 249–256 (2010)
Zeiler, M.D.: ADADELTA: an adaptive learning rate method. Computer Science (2012)
Acknowledgments
This work was supported in part by the University Innovation Team Project of Jinan (2019GXRC015), and in part by Key Science &Technology Innovation Project of Shandong Province (2019JZZY010324), the Natural Science Foundation of China (No. 61902337), Natural Science Fund for Colleges and Universities in Jiangsu Prov-ince (No. 19KJB520016), Jiangsu Provincial Natural Science Foundation (No. SBK2019040953), Young talents of science and technology in Jiangsu.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Zhang, Y., Chen, Y., Bao, W., Cao, Y. (2021). A Hybrid Deep Neural Network for the Prediction of In-Vivo Protein-DNA Binding by Combining Multiple-Instance Learning. In: Huang, DS., Jo, KH., Li, J., Gribova, V., Premaratne, P. (eds) Intelligent Computing Theories and Application. ICIC 2021. Lecture Notes in Computer Science(), vol 12838. Springer, Cham. https://doi.org/10.1007/978-3-030-84532-2_34
Download citation
DOI: https://doi.org/10.1007/978-3-030-84532-2_34
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-84531-5
Online ISBN: 978-3-030-84532-2
eBook Packages: Computer ScienceComputer Science (R0)