Protein 8-class secondary structure prediction using conditional neural fields

By Zhiyong Wang, Feng Zhao, Jian Peng, Jinbo Xu. 2010

130 citations recorded by the source corpus.

Compared to the protein 3-class secondary structure (SS) prediction, the 8-class prediction gains less attention and is also much more challenging, especially for proteins with few sequence homologs. This paper presents a new probabilistic method for 8-class SS prediction using Conditional Neural Fields (CNFs), a recently-invented probabilistic graphical model. This CNF method not only models complex relationship between sequence features and SS, but also exploits interdependency among SS types of adjacent residues. In addition to sequence profiles, our method also makes use of non-evolutionary information for SS prediction. Tested on the CB513 and RS126 datasets, our method achieves Q8 accuracy 64.9% and 64.7%, respectively, which are much better than the SSpro8 web server (51.0% and 48.0%, respectively). Our method can also be used to predict other structure properties (e.g., solvent accessibility) of a protein or the SS of RNA.

Subfields: Materials Chemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Materials Chemistry

1. Jinbo Xu
2. Sheng Wang
3. Yaoqi Zhou
4. Lukasz Kurgan
5. Jianlin Cheng
6. Kuldip K. Paliwal
7. Jack Hanson
8. Jianzhu Ma

In Molecular Biology

1. Jinbo Xu
2. Sheng Wang
3. Jianlin Cheng
4. Yaoqi Zhou
5. Kuldip K. Paliwal
6. Jack Hanson
7. Jianzhu Ma
8. Lukasz Kurgan

Citations in the corpus

• Deep Supervised and Convolutional Generative Stochastic Network for Protein Secondary Structure Prediction (2014, methodology): “Our model achieves Q8 accuracy of 0.664, outperforming previous state-of-the-art result by CNF/RaptorSS8 (0.649) (Wang et al., 2011).”
• A Deep Learning Network Approach to ab initio Protein Secondary Structure Prediction (2014, methodology): “DNSS 81.1 74.7 80.2 73.6 80.7 74.2 50.4 46.1 PSSpred 83.3 72.0 81.0 70.4 82.2 71.3 41.8 33.7 SSpro 79.6 72.6 78.8 71.9 79.2 72.3 49.6 34.0 PSIPRED 80.9 69.3 81.2 68.6 81.0 69.0 33.8 23.2 RaptorX 78.1 70.4 77.9 70.3 78.0 70.3 45.6 33.0 Scores show the average accuracy of secondary structure predic”
• A position-specific distance-dependent statistical potential for protein structure and functional study (2012, methodology): “Evolutionary information has been extensively used in protein secondary structure prediction(Jones, 1999; Wang et al., 2011), fold recognition (Maiorov and Crippen, 1992; Panchenko et al., 2000; Peng and Xu, 2009, 2010; Sippl and Weitckus, 1992; Skolnick et al., 2000), protein alignment (Notredame…”
• Protein Secondary Structure Prediction with Long Short Term Memory Networks (2014, methodology): “Recent work includes conditional random fields hybrid models [Maaten et al., 2011; Peng et al., 2009; Wang et al., 2011] and generative stochastic networks [Troyanskaya, 2014].”
• AUC-Maximized Deep Convolutional Neural Fields for Protein Sequence Labeling (2016, methodology): “Some SS labels, such as 3–10 helix, beta-bridge, and pi-helix are extremely rare [46].”
• Protein Secondary Structure Prediction Using Cascaded Convolutional and Recurrent Neural Networks (2016, methodology): “We also compare our model against other existing methods (e.g., CNF [Wang et al. , 2011], SSpro 8 [Pollastri et al. , 2002] and GSN [Zhou and Troyan-skaya, 2014]) on individual secondary structure labels in Table 2.”
• Sixty-five years of the long march in protein secondary structure prediction: the final stretch? (2016, methodology): “This leads to several recent methods dedicated to eight-state prediction such as SSpro8 [121], RaptorXss8 [122], SCORPION [96, 123] and DeepCNF [58].”
• Deep learning methods in protein structure prediction (2020, background): “The most recent version of RaptorX-Property is based on Deep Convolutional Neural Fields (DeepCNF), i.e. CNN with CRF output [80,23].”
• DeepCNF-D: Predicting Protein Order/Disorder Regions by Weighted Deep Convolutional Neural Fields (2015, methodology): “Here we use the predicted eight-state secondary structure elements (SSE) [18] and three-state solvent accessibility (ACC) [21] probability as structure related features for each residue position.”