Improved protein structure prediction by deep learning irrespective of co-evolution information

By Jinbo Xu, Matthew McPartlon, Jin Li. 2020

210 citations recorded by the source corpus.

Predicting the tertiary structure of a protein from its primary sequence has been greatly improved by integrating deep learning and co-evolutionary analysis, as shown in CASP13 and CASP14. We describe our latest study of this idea, analysing the efficacy of network size and co-evolution data and its performance on both natural and designed proteins. We show that a large ResNet (convolutional residual neural networks) can predict structures of correct folds for 26 out of 32 CASP13 free-modelling targets and L/5 long-range contacts with precision over 80%. When co-evolution is not used, ResNet can still predict structures of correct folds for 18 CASP13 free-modelling targets, greatly exceeding previous methods that do not use co-evolution either. Even with only the primary sequence, ResNet can predict the structures of correct folds for all tested human-designed proteins. In addition, ResNet may fare better for the designed proteins when trained without co-evolution than with co-evolution. These results suggest that ResNet does not simply de-noise co-evolution signals, but instead may learn important protein sequence–structure relationships. This has important implications for protein design and engineering, especially when co-evolutionary data are unavailable. In the last few years, computational protein structure prediction has greatly advanced by combining deep learning including convolutional residual networks (ResNet) with co-evolution data. A new study finds that using deeper and wider ResNets improves predictions in the absence of co-evolution information, suggesting that the ResNets do not not simply de-noise co-evolution signals, but instead may learn important protein sequence–structure relationships.

Subfields: Molecular Biology, Computational Biology & Bioinformatics

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

In Computational Biology & Bioinformatics

Citations in the corpus

Single-sequence protein structure prediction using language models from deep learning (2021, unknown): “48 .”
A guide to machine learning for biologists (2021, background): “In this case, the network learns to pick out direct coupling interactions, and accurate predictions can be made even for sequences with few or no related sequences [56].”
OpenFold: retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization (2023, unknown): “Building on a line of work applying deep learning to coevolutionary information encoded in multiple-sequence alignments (MSAs) 3–8 and homologous structures 9,10 , AlphaFold2 (ref.”
Single-sequence protein structure prediction using supervised transformer protein language models (2022, background): “The remaining targets are then merged with the 35 designed proteins from previous works 3, 21 , resulting in 101 human-designed targets.”
Using AlphaFold for Rapid and Accurate Fixed Backbone Protein Design (2021, background): “Protein structure prediction methods tend to perform well on designed proteins in the absence of coevolutionary information (Xu et al., 2021) and it is reasonable to expect that within AlphaFold is the potential for rapid and accurate protein design.”
Single-sequence protein structure prediction using language models and deep learning (2022, methodology): “Performance of RGN2 was compared against trRosetta across two sets of non-homologous proteins: (1) 129 orphans from the Uniclust30 database 51 and (2) 35 de novo proteins by Xu et al. 52 .”