Protein secondary structure prediction using deep convolutional neural fields

TL;DR

This paper introduces DeepCNF, a deep learning model that significantly improves protein secondary structure prediction accuracy by modeling complex sequence-structure relationships and label interdependencies.

Contribution

DeepCNF extends Conditional Neural Fields with deep hierarchical architecture, achieving higher accuracy and broader applicability in protein structure prediction tasks.

Findings

Achieved ~84% Q3 accuracy on test proteins

Outperformed existing predictors in accuracy and SOV score

Can be applied to predict other protein structure properties

Abstract

Protein secondary structure (SS) prediction is important for studying protein structure and function. When only the sequence (profile) information is used as input feature, currently the best predictors can obtain ~80% Q3 accuracy, which has not been improved in the past decade. Here we present DeepCNF (Deep Convolutional Neural Fields) for protein SS prediction. DeepCNF is a Deep Learning extension of Conditional Neural Fields (CNF), which is an integration of Conditional Random Fields (CRF) and shallow neural networks. DeepCNF can model not only complex sequence-structure relationship by a deep hierarchical architecture, but also interdependency between adjacent SS labels, so it is much more powerful than CNF. Experimental results show that DeepCNF can obtain ~84% Q3 accuracy, ~85% SOV score, and ~72% Q8 accuracy, respectively, on the CASP and CAMEO test proteins, greatly outperforming currently popular predictors. As a general framework, DeepCNF can be used to predict other protein structure properties such as contact number, disorder regions, and solvent accessibility.