Analysis of deep learning methods for blind protein contact prediction in CASP12

By Sheng Wang, Siqi Sun, Jinbo Xu. 2017

90 citations recorded by the source corpus.

Here we present the results of protein contact prediction achieved in CASP12 by our RaptorX-Contact server, which is an early implementation of our deep learning method for contact prediction. On a set of 38 free-modeling target domains with a median family size of around 58 effective sequences, our server obtained an average top L/5 long- and medium-range contact accuracy of 47% and 44%, respectively (L=length). A more advanced implementation has an average accuracy of 59% and 57%, respectively. Our deep learning method formulates contact prediction as an image pixel-level labeling problem and simultaneously predicts all residue pairs of a protein using a combination of two deep residual neural networks, taking as input the residue conservation information, predicted secondary structure and solvent accessibility, contact potential, and co-evolution information. Our approach differs from existing methods mainly in (1) formulating contact prediction as a pixel-level image labeling problem instead of an image-level classification problem; (2) simultaneously predicting all contacts of an individual protein to make effective use of contact occurrence patterns; and (3) integrating both 1D and 2D deep convolutional neural networks to effectively learn complex sequence-structure relationship including high-order residue correlation. This paper discusses the RaptorX-Contact pipeline, both contact prediction and contact-based folding results, and finally the strength and weakness of our method.

Subfields: 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 Molecular Biology

Citations in the corpus

Critical Assessment of Methods of Protein Structure Prediction (CASP) – Round XIII (2019, background): “instance.(10,11) But the key algorithmic advance appears to be the”
Assessment of Contact Predictions in CASP12: Co-Evolution and Deep Learning Coming of Age (2017, methodology): “Some predictors indicate that improved methodology including deeper networks is one of the main drivers for the significant improvement in predictions.(48,49) Another reason for the enhanced performance in CASP12 appears to be the increase in the number of sequences available (Figure 16) and, thus, ”
ComplexContact: a web server for inter-protein contact prediction using deep learning (2018, methodology): “The underlying algorithm of this server is a Deep Learning (DL) model which has won intra-protein contact prediction in CASP12 (24,25).”
Recent developments in deep learning applied to protein structure prediction (2019, methodology): “Recurrent architectures have also been used in contact prediction.(48,89) More recently, a recurrent architecture has been used to model tertiary structure.”
De novo Protein Structure Prediction by Coupling Contact With Distance Profile (2020, methodology): “( 2): where CS 1 m;n is the contact conﬁdence score obtained by RaptorX-Contact [18], [26], [27], [28], [60] for the m th and n th residues, CS 2 m;n is another contact conﬁdence score obtained by ResPRE [25] for the m th and n th residues.”