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

arXiv: https://arxiv.org/abs/1403.1347 · HTML: https://mlsanity.com/paper/1403.1347/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Mar 2014.

## Built on by (most cited first, 8 of 8)

- [Protein Language Models and Structure Prediction: Connection and Progression](https://mlsanity.com/paper/2211.16742/) (2022)
- [Designing ReLU Generative Networks to Enumerate Trees with a Given Tree Edit Distance](https://mlsanity.com/paper/2510.10706/) (2025)
- [Adaptive Residue-wise Profile Fusion for Low Homologous Protein SecondaryStructure Prediction Using External Knowledge](https://mlsanity.com/paper/2108.04176/) (2021)
- [Learning Sequence Representations by Non-local Recurrent Neural Memory](https://mlsanity.com/paper/2207.09710/) (2022)
- [Protein Secondary Structure Prediction Using 3D Graphs and Relation-Aware Message Passing Transformers](https://mlsanity.com/paper/2511.13685/) (2025)
- [Multi-objective Genetic Programming with Multi-view Multi-level Feature for Enhanced Protein Secondary Structure Prediction](https://mlsanity.com/paper/2603.12293/) (2026)
- [The limits of bio-molecular modeling with large language models : a cross-scale evaluation](https://mlsanity.com/paper/2604.03361/) (2026)
- [ReLU Networks for Exact Generation of Similar Graphs](https://mlsanity.com/paper/2604.05929/) (2026)

## Abstract (arXiv)

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.