Deep learning reveals many more inter-protein residue-residue contacts than direct coupling analysis

Zhou, Tianming; Wang, Sheng; Xu, Jinbo

doi:10.1101/240754

Cited by 24 publications

(42 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The detailed description of the DL algorithm underlying ComplexContact is described in ( 16 ) and the detailed experimental results for inter-protein contact prediction is available at ( 26 ). Here, we briefly summarize the method.…”

Section: Methodsmentioning

confidence: 99%

ComplexContact: a web server for inter-protein contact prediction using deep learning

Zeng

Wang

Zhou

et al. 2018

Nucleic Acids Research

Self Cite

135

142

View full text Add to dashboard Cite

ComplexContact (http://raptorx2.uchicago.edu/ComplexContact/) is a web server for sequence-based interfacial residue-residue contact prediction of a putative protein complex. Interfacial residue-residue contacts are critical for understanding how proteins form complex and interact at residue level. When receiving a pair of protein sequences, ComplexContact first searches for their sequence homologs and builds two paired multiple sequence alignments (MSA), then it applies co-evolution analysis and a CASP-winning deep learning (DL) method to predict interfacial contacts from paired MSAs and visualizes the prediction as an image. The DL method was originally developed for intra-protein contact prediction and performed the best in CASP12. Our large-scale experimental test further shows that ComplexContact greatly outperforms pure co-evolution methods for inter-protein contact prediction, regardless of the species.

show abstract

Section: Methodsmentioning

confidence: 99%

ComplexContact: a web server for inter-protein contact prediction using deep learning

Zeng

Wang

Zhou

et al. 2018

Nucleic Acids Research

Self Cite

135

142

View full text Add to dashboard Cite

show abstract

“…We also defined interchain contact between chains in a protein if the Euclidean distance between the heavy atoms of the residues in the respective chains is less than or equal to 6.0 Å (Hopf et al, 2014;Ovchinnikov et al, 2014;Zhou et al, 2018). We obtained a pairwise contact list between the first chain (chain A) and the other chains after separating the homomultimeric PDB into separate chains.…”

Section: Prediction and Evaluation Of Interchain Contactsmentioning

confidence: 99%

“…Machine learning methods have been developed to facilitate computational modeling of both protein tertiary structures and quaternary structures. However, most of the recent focus has been on the development of computational tools for predicting intrachain (within the same chain) residue-residue contacts and distances to guide tertiary structure modeling (Hopf et al, 2014;Zhou et al, 2018). Some of these methods have performed well in the 12 th and 13 th Critical Assessment of Techniques for Protein Structure Prediction (CASP) competitions (Adhikari et al, 2018;Alquraishi & Valencia, 2019;Cheng et al, 2019;Schaarschmidt et al, 2018;Senior et al, 2020;Shrestha, Fajardo, Gil, Fidelis, Kryshtafovych, Bohdan Monastyrskyy, et al, 2019;Wang et al, 2017Xu & Wang, 2019).…”

Section: Introductionmentioning

confidence: 99%

“…Unlike tertiary structure modeling, most of the tools on protein complexes are developed to classify whether two proteins are in contact or not, with some tools developed exclusively for docking tertiary structures together. Only a small number of tools were developed to predict interchain contacts leveraging interchain residue-residue co-evolutionary signals embedded in the multiple sequence alignments of a pair of interacting proteins (i.e., interlog) (Hopf et al, 2014;Ovchinnikov et al, 2014;Zhou et al, 2018). Relevant to the present work (Zhou et al, 2018) and (Zeng et al, 2018) predicted interchain contacts using a deep learning-based intrachain contact prediction tool (RaptorX-ComplexContact) without training the system for interchain contact prediction.…”

Section: Introductionmentioning

confidence: 99%

“…Only a small number of tools were developed to predict interchain contacts leveraging interchain residue-residue co-evolutionary signals embedded in the multiple sequence alignments of a pair of interacting proteins (i.e., interlog) (Hopf et al, 2014;Ovchinnikov et al, 2014;Zhou et al, 2018). Relevant to the present work (Zhou et al, 2018) and (Zeng et al, 2018) predicted interchain contacts using a deep learning-based intrachain contact prediction tool (RaptorX-ComplexContact) without training the system for interchain contact prediction. Their work involved generating MSAs using homology-based, phylogeny-based, and genome-based interlog (interacting homologs of protein complexes) searches, which is similar to methods employed by (Ovchinnikov et al, 2014) and (Hopf et al, 2014) to generate MSA.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Predicting interchain contacts for homodimeric and homomultimeric protein complexes using multiple sequence alignments of monomers and deep learning

Quadir

Roy

Halfmann

et al. 2020

Preprint

View full text Add to dashboard Cite

Deep learning methods that achieved great success in predicting intrachain residue-residue contacts have been applied to predict interchain contacts between proteins. However, these methods require multiple sequence alignments (MSAs) of a pair of interacting proteins (dimers) as input, which are often difficult to obtain because there are not many known protein complexes available to generate MSAs of sufficient depth for a pair of proteins. In recognizing that multiple sequence alignments of a monomer that forms homomultimers contain the co-evolutionary signals of both intrachain and interchain residue pairs in contact, we applied DNCON2 (a deep learning-based protein intrachain residue-residue contact predictor) to predict both intrachain and interchain contacts for homomultimers using multiple sequence alignment (MSA) and other co-evolutionary features of a single monomer followed by discrimination of interchain and intrachain contacts according to the tertiary structure of the monomer. Allowing true-positive predictions within two residue shifts, the best average precision was obtained for the Top-L/10 predictions of DNCON2: 22.9% for homodimers, and 17.0% for higher order homomultimers. In some instances, especially where interchain contact densities are high, the approach predicted interchain contacts with 100% precision. We show that the predicted contacts can be used to accurately construct the structure of some complexes. Our experiment demonstrates that monomeric multiple sequence alignments can be used with deep learning to predict interchain contacts of homomeric proteins.

show abstract

Distance‐based reconstruction of protein quaternary structures from inter‐chain contacts

et al. 2021

View full text Add to dashboard Cite

Predicting the quaternary structure of protein complex is an important problem. Interchain residue-residue contact prediction can provide useful information to guide the ab initio reconstruction of quaternary structures. However, few methods have been developed to build quaternary structures from predicted inter-chain contacts. Here, we develop the first method based on gradient descent optimization (GD) to build quaternary structures of protein dimers utilizing inter-chain contacts as distance restraints. We evaluate GD on several datasets of homodimers and heterodimers using true/predicted contacts and monomer structures as input. GD consistently performs better than both simulated annealing and Markov Chain Monte Carlo simulation. Starting from an arbitrarily quaternary structure randomly initialized from the tertiary structures of protein chains and using true inter-chain contacts as input, GD can reconstruct high-quality structural models for homodimers and heterodimers with average TM-score ranging from 0.92 to 0.99 and average interface root mean square distance from 0.72 Å to 1.64 Å. On a dataset of 115 homodimers, using predicted inter-chain contacts as restraints, the average TM-score of the structural models built by GD is 0.76. For 46% of the homodimers, high-quality structural models with TM-score ≥ 0.9 are reconstructed from predicted contacts. There is a strong correlation between the quality of the reconstructed models and the precision and recall of predicted contacts. Only a moderate precision or recall of inter-chain contact prediction is needed to build good structural models for most homodimers. Moreover, GD improves the quality of quaternary structures predicted by AlphaFold2 on a Critical Assessment of Techniques for Protein Structure Prediction-Critical Assessments of Predictions of Interactions dataset.

show abstract

Deep learning reveals many more inter-protein residue-residue contacts than direct coupling analysis

Cited by 24 publications

References 57 publications

ComplexContact: a web server for inter-protein contact prediction using deep learning

ComplexContact: a web server for inter-protein contact prediction using deep learning

Predicting interchain contacts for homodimeric and homomultimeric protein complexes using multiple sequence alignments of monomers and deep learning

Distance‐based reconstruction of protein quaternary structures from inter‐chain contacts

Contact Info

Product

Resources

About