Deep Supervised and Convolutional Generative Stochastic Network for Protein Secondary Structure Prediction

Jian Zhou, Olga Troyanskaya
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(1):745-753, 2014.

Abstract

Predicting protein secondary structure is a fundamental problem in protein structure prediction. Here we present a new supervised generative stochastic network (GSN) based method to predict local secondary structure with deep hierarchical representations. GSN is a recently proposed deep learning technique (Bengio & Thibodeau-Laufer, 2013) to globally train deep generative model. We present the supervised extension of GSN, which learns a Markov chain to sample from a conditional distribution, and applied it to protein structure prediction. To scale the model to full-sized, high-dimensional data, like protein sequences with hundreds of amino-acids, we introduce a convolutional architecture, which allows efficient learning across multiple layers of hierarchical representations. Our architecture uniquely focuses on predicting structured low-level labels informed with both low and high-level representations learned by the model. In our application this corresponds to labeling the secondary structure state of each amino-acid residue. We trained and tested the model on separate sets of non-homologous proteins sharing less than 30% sequence identity. Our model achieves 66.4% Q8 accuracy on the CB513 dataset, better than the previously reported best performance 64.9% (Wang et al., 2011) for this challenging secondary structure prediction problem.

Cite this Paper

BibTeX
@InProceedings{pmlr-v32-zhou14, title = {Deep Supervised and Convolutional Generative Stochastic Network for Protein Secondary Structure Prediction}, author = {Zhou, Jian and Troyanskaya, Olga}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {745--753}, year = {2014}, editor = {Xing, Eric P. and Jebara, Tony}, volume = {32}, number = {1}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/zhou14.pdf}, url = {https://proceedings.mlr.press/v32/zhou14.html}, abstract = {Predicting protein secondary structure is a fundamental problem in protein structure prediction. Here we present a new supervised generative stochastic network (GSN) based method to predict local secondary structure with deep hierarchical representations. GSN is a recently proposed deep learning technique (Bengio & Thibodeau-Laufer, 2013) to globally train deep generative model. We present the supervised extension of GSN, which learns a Markov chain to sample from a conditional distribution, and applied it to protein structure prediction. To scale the model to full-sized, high-dimensional data, like protein sequences with hundreds of amino-acids, we introduce a convolutional architecture, which allows efficient learning across multiple layers of hierarchical representations. Our architecture uniquely focuses on predicting structured low-level labels informed with both low and high-level representations learned by the model. In our application this corresponds to labeling the secondary structure state of each amino-acid residue. We trained and tested the model on separate sets of non-homologous proteins sharing less than 30% sequence identity. Our model achieves 66.4% Q8 accuracy on the CB513 dataset, better than the previously reported best performance 64.9% (Wang et al., 2011) for this challenging secondary structure prediction problem.} }
Endnote
%0 Conference Paper %T Deep Supervised and Convolutional Generative Stochastic Network for Protein Secondary Structure Prediction %A Jian Zhou %A Olga Troyanskaya %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-zhou14 %I PMLR %P 745--753 %U https://proceedings.mlr.press/v32/zhou14.html %V 32 %N 1 %X Predicting protein secondary structure is a fundamental problem in protein structure prediction. Here we present a new supervised generative stochastic network (GSN) based method to predict local secondary structure with deep hierarchical representations. GSN is a recently proposed deep learning technique (Bengio & Thibodeau-Laufer, 2013) to globally train deep generative model. We present the supervised extension of GSN, which learns a Markov chain to sample from a conditional distribution, and applied it to protein structure prediction. To scale the model to full-sized, high-dimensional data, like protein sequences with hundreds of amino-acids, we introduce a convolutional architecture, which allows efficient learning across multiple layers of hierarchical representations. Our architecture uniquely focuses on predicting structured low-level labels informed with both low and high-level representations learned by the model. In our application this corresponds to labeling the secondary structure state of each amino-acid residue. We trained and tested the model on separate sets of non-homologous proteins sharing less than 30% sequence identity. Our model achieves 66.4% Q8 accuracy on the CB513 dataset, better than the previously reported best performance 64.9% (Wang et al., 2011) for this challenging secondary structure prediction problem.
RIS
TY - CPAPER TI - Deep Supervised and Convolutional Generative Stochastic Network for Protein Secondary Structure Prediction AU - Jian Zhou AU - Olga Troyanskaya BT - Proceedings of the 31st International Conference on Machine Learning DA - 2014/01/27 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-zhou14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 32 IS - 1 SP - 745 EP - 753 L1 - http://proceedings.mlr.press/v32/zhou14.pdf UR - https://proceedings.mlr.press/v32/zhou14.html AB - Predicting protein secondary structure is a fundamental problem in protein structure prediction. Here we present a new supervised generative stochastic network (GSN) based method to predict local secondary structure with deep hierarchical representations. GSN is a recently proposed deep learning technique (Bengio & Thibodeau-Laufer, 2013) to globally train deep generative model. We present the supervised extension of GSN, which learns a Markov chain to sample from a conditional distribution, and applied it to protein structure prediction. To scale the model to full-sized, high-dimensional data, like protein sequences with hundreds of amino-acids, we introduce a convolutional architecture, which allows efficient learning across multiple layers of hierarchical representations. Our architecture uniquely focuses on predicting structured low-level labels informed with both low and high-level representations learned by the model. In our application this corresponds to labeling the secondary structure state of each amino-acid residue. We trained and tested the model on separate sets of non-homologous proteins sharing less than 30% sequence identity. Our model achieves 66.4% Q8 accuracy on the CB513 dataset, better than the previously reported best performance 64.9% (Wang et al., 2011) for this challenging secondary structure prediction problem. ER -
APA
Zhou, J. & Troyanskaya, O.. (2014). Deep Supervised and Convolutional Generative Stochastic Network for Protein Secondary Structure Prediction. Proceedings of the 31st International Conference on Machine Learning, in Proceedings of Machine Learning Research 32(1):745-753 Available from https://proceedings.mlr.press/v32/zhou14.html.

Related Material