Large-scale model quality assessment for improving protein tertiary structure prediction

Cao, Renzhi; Bhattacharya, Debswapna; Adhikari, Badri; Li, Jilong; Cheng, Jianlin

doi:10.1093/bioinformatics/btv235

Cited by 60 publications

(56 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[8][9][10][11][12] The improved contact prediction led to the significant improvement of template-FM in CASP12 experiment, in which contact predictions were used with different ab initio modeling methods such as fragment assembly and distance geometry to build protein structural models from scratch. 1 To prepare for 2018 CASP13 experiment, we focused on enhancing our MULTICOM protein structure prediction system [17][18][19] with our latest development in contact distance prediction empowered by deep learning and its application to template-FM and protein model ranking, 17,20,21 while having a routine update on its other components such as template library, template identification, and template-based modeling. Our experiment demonstrates that contact distance prediction empowered by the advanced deep learning architecture can accurately predict a large number of contacts for some template-free or hard template-based targets, which are sufficient to fold them correctly by the distance geometry and simulated annealing from scratch without using any template or fragment information.…”

Section: Introductionmentioning

confidence: 99%

Protein tertiary structure modeling driven by deep learning and contact distance prediction in CASP13

et al. 2019

Self Cite

View full text Add to dashboard Cite

Predicting residue-residue distance relationships (e.g. contacts) has become the key direction to advance protein structure prediction since 2014 CASP11 experiment, while deep learning has revolutionized the technology for contact and distance distribution prediction since its debut in 2012 CASP10 experiment. During 2018 CASP13 experiment, we enhanced our MULTICOM protein structure prediction system with three major components: contact distance prediction based on deep convolutional neural networks, distance-driven template-free (ab initio) modeling, and protein model ranking empowered by deep learning and contact prediction. Our experiment demonstrates that contact distance prediction and deep learning methods are the key reasons that MULTICOM was ranked 3rd out of all 98 predictors in both template-free and template-based structure modeling in CASP13. Deep convolutional neural network can utilize global information in pairwise residue-residue features such as co-evolution scores to substantially improve contact distance prediction, which played a decisive role in correctly folding some free modeling and hard template-based modeling targets. Deep learning also successfully integrated 1D structural features, 2D contact information, and 3D structural quality scores to improve protein model quality assessment, where the contact prediction was demonstrated to consistently enhance ranking of protein models for the first time. The success of MULTICOM system clearly shows that protein contact distance prediction and model selection driven by deep learning holds the key of solving protein structure prediction problem. However, there are still challenges in accurately predicting protein contact distance when there are few homologous sequences, folding proteins from noisy contact distances, and ranking models of hard targets.

show abstract

Section: Introductionmentioning

confidence: 99%

Protein tertiary structure modeling driven by deep learning and contact distance prediction in CASP13

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Estimating the quality of a computationally generated protein structural model serves as a key component of protein structure prediction (Won et al, 2019;Uziela et al, 2017). Model quality estimation assists in validating and evaluating predicted protein models at multiple stages of a structure prediction pipeline, thus greatly affecting its prediction accuracy (Cao et al, 2015;Kalman and Ben-Tal, 2010). Methods for model quality estimation can be broadly categorized into two major classes that include "single-model" methods and "consensus" approaches.…”

Section: Introductionmentioning

confidence: 99%

QDeep: distance-based protein model quality estimation by residue-level ensemble error classifications using stacked deep residual neural networks

Shuvo

Bhattacharya

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Motivation:Protein model quality estimation, in many ways, informs protein structure prediction. Despite their tight coupling, existing model quality estimation methods do not leverage inter-residue distance information or the latest technological breakthrough in deep learning that has recently revolutionized protein structure prediction. Results: We present a new distance-based single-model quality estimation method called QDeep by harnessing the power of stacked deep residual neural networks (ResNets). Our method first employs stacked deep ResNets to perform residue-level ensemble error classifications at multiple predefined error thresholds, and then combines the predictions from the individual error classifiers for estimating the quality of a protein structural model. Experimental results show that our method consistently outperforms existing state-of-the-art methods including ProQ2, ProQ3, ProQ3D, ProQ4, 3DCNN, MESHI, and VoroMQA in multiple independent test datasets across a wide-range of accuracy measures; and that predicted distance information significantly contributes to the improved performance of QDeep.

show abstract

“…As a quality control step of modeling, protein model quality assessment (QA) plays an important role in selecting most accurate models among a massive number of decoys generated by protein structure modeling methods. There are two kinds of model quality assessment methods: local quality assessment [8][9][10][11][12] and global quality assessment 10,[13][14][15][16][17][18][19][20][21][22] . Local QA methods attempt to predict the spatial deviation of each residue in a model from the native structure (e.g.…”

Section: Introductionmentioning

confidence: 99%

“…Several features have been proved to be effective, such as sequence/profile alignment, predicted secondary structure and solvent accessibility of residues 8 , residue-residue contact potential 19 , torsion angle of main chain 27 , physicochemical properties 13 , and energy-based environment of residues and models 12,13,15 . Methods such as support vector machine 8,15,28 , neural network 14 , and linear combination 22,29 are commonly used for quality estimation. Many top QA methods have been largely tested and assessed in the Critical Assessment of Techniques for Protein Structure Prediction (CASP) 9 .…”

Section: Introductionmentioning

confidence: 99%

Deep convolutional neural networks for predicting the quality of single protein structural models

Hou

Cao

Cheng

2019

Preprint

Self Cite

View full text Add to dashboard Cite

Predicting the global quality and local (residual-specific) quality of a single protein structural model is important for protein structure prediction and application. In this work, we developed a deep one-dimensional convolutional neural network (1DCNN) that predicts the absolute local quality of a single protein model as well as two 1DCNNs to predict both local and global quality simultaneously through a novel multi-task learning framework. The networks accept sequential and structural features (i.e. amino acid sequence, agreement of secondary structure and solvent accessibilities, residual disorder properties and Rosetta energies) of a protein model of any size as input to predict its quality, which is different from existing methods using a fixed number of handcrafted features as input. Our three methods (InteractQA-net, JointQA-net and LocalQA-net) were trained on the structural models of the single-domain protein targets of CASP8, 9, 10 and evaluated on the models of CASP11 and CASP12 targets. The results show that the performance of our deep learning methods is comparable to the state-of-the-art quality assessment methods. Our study also demonstrates that combining local and global quality predictions together improves the global 2 quality prediction accuracy. The source code and executable of our methods are available at: https://github.com/multicom-toolbox/DeepCovQA Affiliations

show abstract

Large-scale model quality assessment for improving protein tertiary structure prediction

Cited by 60 publications

References 42 publications

Protein tertiary structure modeling driven by deep learning and contact distance prediction in CASP13

Protein tertiary structure modeling driven by deep learning and contact distance prediction in CASP13

QDeep: distance-based protein model quality estimation by residue-level ensemble error classifications using stacked deep residual neural networks

Deep convolutional neural networks for predicting the quality of single protein structural models

Contact Info

Product

Resources

About