Search Menu
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information
Search Scholarly Publications

Title: Combining pairwise structural similarity and deep learning interface contact prediction to estimate protein complex model accuracy in CASP15

Journal Article · · Proteins
DOI: https://doi.org/10.1002/prot.26542 · OSTI ID:1986586
 [1];  [1];  [1];  [1]; ORCiD logo [1]
  1. Department of Electrical Engineering and Computer Science, NextGen Precision Health University of Missouri Columbia Missouri USA
+ Show Author Affiliations

Abstract Estimating the accuracy of quaternary structural models of protein complexes and assemblies (EMA) is important for predicting quaternary structures and applying them to studying protein function and interaction. The pairwise similarity between structural models is proven useful for estimating the quality of protein tertiary structural models, but it has been rarely applied to predicting the quality of quaternary structural models. Moreover, the pairwise similarity approach often fails when many structural models are of low quality and similar to each other. To address the gap, we developed a hybrid method (MULTICOM_qa) combining a pairwise similarity score (PSS) and an interface contact probability score (ICPS) based on the deep learning inter‐chain contact prediction for estimating protein complex model accuracy. It blindly participated in the 15th Critical Assessment of Techniques for Protein Structure Prediction (CASP15) in 2022 and performed very well in estimating the global structure accuracy of assembly models. The average per‐target correlation coefficient between the model quality scores predicted by MULTICOM_qa and the true quality scores of the models of CASP15 assembly targets is 0.66. The average per‐target ranking loss in using the predicted quality scores to rank the models is 0.14. It was able to select good models for most targets. Moreover, several key factors (i.e., target difficulty, model sampling difficulty, skewness of model quality, and similarity between good/bad models) for EMA are identified and analyzed. The results demonstrate that combining the multi‐model method (PSS) with the complementary single‐model method (ICPS) is a promising approach to EMA.

Research Organization:
Donald Danforth Plant Science Center, St. Louis, MO (United States)
Sponsoring Organization:
National Institutes of Health (NIH); National Science Foundation (NSF); USDOE
Grant/Contract Number:
SC0020400; SC0021303
OSTI ID:
1986586
Journal Information:
Proteins, Journal Name: Proteins Journal Issue: 12 Vol. 91; ISSN 0887-3585
Publisher:
Wiley Blackwell (John Wiley & Sons)Copyright Statement
Country of Publication:
United States
Language:
English

References (45)

SPICKER: A clustering approach to identify near-native protein folds journal January 2004
ZDOCK: An initial-stage protein-docking algorithm journal May 2003
Scoring function for automated assessment of protein structure template quality journal January 2004
ZRANK: Reranking protein docking predictions with an optimized energy function journal March 2007
A combination of rescoring and refinement significantly improves protein docking performance: Rescoring and Refinement in Protein Docking journal January 2008
Assessment of global and local model quality in CASP8 using Pcons and ProQ: Pcons and ProQ MQAPs in CASP8 journal January 2009
Prediction of global and local quality of CASP8 models by MULTICOM series: Prediction of Quality of CASP8 Model journal January 2009
CASP prediction center infrastructure and evaluation measures in CASP10 and CASP ROLL: CASP Prediction Center journal October 2013
Prediction of homoprotein and heteroprotein complexes by protein docking and template‐based modeling: A CASP‐CAPRI experiment journal June 2016
Critical assessment of methods of protein structure prediction: Progress and new directions in round XI: Progress in CASP XI journal June 2016
VoroMQA: Assessment of protein structure quality using interatomic contact areas: Contact Area-based Protein Structure Assessment journal March 2017
The challenge of modeling protein assemblies: the CASP12-CAPRI experiment journal November 2017
Protein tertiary structure modeling driven by deep learning and contact distance prediction in CASP13 journal April 2019
Estimation of model accuracy in CASP13 journal July 2019
Prediction of interresidue contacts with DeepMetaPSICOV in CASP13 journal July 2019
Deep‐learning contact‐map guided protein structure prediction in CASP13 journal August 2019
Critical assessment of methods of protein structure prediction (CASP)—Round XIII journal August 2019
Protein structure prediction using multiple deep neural networks in the 13th Critical Assessment of Protein Structure Prediction (CASP13) journal August 2019
Modeling of protein complexes in CASP14 with emphasis on the interaction interface prediction journal July 2021
Improving protein tertiary structure prediction by deep learning and distance prediction in CASP14 journal July 2021
Assessment of protein model structure accuracy estimation in CASP14 : Old and new challenges journal August 2021
Prediction of protein assemblies, the next frontier: The CASP14‐CAPRI experiment journal September 2021
A large-scale experiment to assess protein structure prediction methods journal November 1995
Development of Unified Statistical Potentials Describing Protein-Protein Interactions journal March 2003
Highly accurate protein structure prediction with AlphaFold journal July 2021
Improved protein structure prediction using predicted interresidue orientations journal January 2020
Accurate prediction of inter-protein residue–residue contacts for homo-oligomeric protein complexes journal March 2021
Improved model quality assessment using sequence and structural information by enhanced deep neural networks journal December 2022
VoroCNN: deep convolutional neural network built on 3D Voronoi tessellation of protein structures journal February 2021
Deep graph learning of inter-protein contacts journal November 2021
A deep dilated convolutional residual network for predicting interchain contacts of protein homodimers journal February 2022
3D-equivariant graph neural networks for protein model quality assessment journal January 2023
APOLLO: a quality assessment service for single and multiple protein models journal May 2011
Predicting protein residue–residue contacts using deep networks and boosting journal October 2012
Protein docking model evaluation by 3D deep convolutional neural networks journal November 2019
ModFOLD8: accurate global and local quality estimates for 3D protein models journal May 2021
LGA: a method for finding 3D similarities in protein structures journal July 2003
MM-align: a quick algorithm for aligning multiple-chain protein complex structures using iterative dynamic programming journal May 2009
Protein complex prediction with AlphaFold-Multimer preprint March 2022
Identification of correct regions in protein models using structural, alignment, and consensus information journal March 2006
Accurate prediction of protein structures and interactions using a three-track neural network journal July 2021
Accurate De Novo Prediction of Protein Contact Map by Ultra-Deep Learning Model journal January 2017
DockQ: A Quality Measure for Protein-Protein Docking Models journal August 2016
Protein Docking Model Evaluation by Graph Neural Networks journal May 2021
DeepComplex: A Web Server of Predicting Protein Complex Structures by Deep Learning Inter-chain Contact Prediction and Distance-Based Modelling journal August 2021

Similar Records

Related Subjects

59 BASIC BIOLOGICAL SCIENCES
deep learning
estimation of protein model accuracy
protein interface contact prediction
protein model quality assessment
protein quaternary structure prediction