OPUS-Refine: A Fast Sampling-Based Framework for Refining Protein Backbone Torsion Angles and Global Conformation

Xu, Gang; Wang, Qinghua; Ma, Jianpeng

doi:10.1021/acs.jctc.9b01054

Cited by 4 publications

(8 citation statements)

References 32 publications

(74 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our previous study, we proposed a neighbor-dependent statistical torsion-angle-sampling database, named OPUS-TA, which ensured the accuracy of OPUS-Refine, a torsion angle refinement method. 41 In OPUS-Fold, we continue to use OPUS-TA as the basic sampling database. In addition, the values of torsion angles from the backbone torsion-angleconstraining term and the values of torsion angles corresponding to the input initial structure are also included into the sampling database.…”

Section: ■ Methodsmentioning

confidence: 99%

“…The parameter Coverage is the percentage of all fragments (with length of 5, 7, 9, 11 residues) in a protein that can be found in the entire PDB at least five times. 41 In OPUS-Fold, if the Coverage of a target sequence is less than 0.8 (essentially a low homologous sequence), we use the structure predicted by the RaptorX-Contact server 38 as the initial structure for its good performance on the protein without many sequence homologues. 28 Otherwise, for a higher homologous sequence, we use the structure reconstructed by the SPOT-1D server's predicted torsion angles.…”

Section: ■ Methodsmentioning

confidence: 99%

“…Since OPUS-Fold is based on torsion-angle sampling, an effective torsion-angle-sampling database is necessary. In our previous study, we proposed a neighbor-dependent statistical torsion-angle-sampling database, named OPUS-TA, which ensured the accuracy of OPUS-Refine, a torsion angle refinement method . In OPUS-Fold, we continue to use OPUS-TA as the basic sampling database.…”

Section: Methodsmentioning

confidence: 99%

“…The definition of these three terms can be found in OPUS-Refine. 41 The E LJ term corresponds to the modified Lennard-Jones potential, 36 and it is used to prevent the severe collision between different atoms. The E rota2 term corresponds to the side-chain modeling method OPUS-Rota2, 36 which is especially useful when the backbone is not in the native state.…”

Section: ■ Energy Functionmentioning

confidence: 99%

“…We combine the two test sets used in OPUS-Refine 41 to form the large one used in this article which contains 103 proteins with an average of 85 residues in length.…”

Section: ■ Test Setsmentioning

confidence: 99%

See 4 more Smart Citations

OPUS-Fold: An Open-Source Protein Folding Framework Based on Torsion-Angle Sampling

Wang

2020

J. Chem. Theory Comput.

Self Cite

View full text Add to dashboard Cite

In this article, we propose a protein folding framework, named OPUS-Fold, which can integrate various methods for subproblems in protein structure prediction to contribute to folding. OPUS-Fold is based on torsion-angle sampling. After each sampling step, it reconstructs the structure and estimates the model quality with an energy function that is formed by combining many different constraining terms designed either by ourselves or by others in the literature. OPUS-Fold balances accuracy and efficiency, delivers good results in a short time, and leaves more space for including the results of other subproblem methods. Moreover, OPUS-Fold also contains a fast side-chain modeling method OPUS-Rota2 (J. Chem. Theory Comput. 2019, 15 (9), 5154–5160), which enables a speedy construction of all-atom atomic models during the folding process that allows the usage of all-atom-required subproblem methods. In summary, OPUS-Fold provides a protein folding platform for incorporating the results from various subproblem methods, including those containing nondifferentiable information such as partial experimental data. The source code of OPUS-Fold can be downloaded from .

show abstract

Section: ■ Methodsmentioning

confidence: 99%

Section: ■ Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: ■ Energy Functionmentioning

confidence: 99%

“…We combine the two test sets used in OPUS-Refine 41 to form the large one used in this article which contains 103 proteins with an average of 85 residues in length.…”

Section: ■ Test Setsmentioning

confidence: 99%

See 3 more Smart Citations

OPUS-Fold: An Open-Source Protein Folding Framework Based on Torsion-Angle Sampling

Wang

2020

J. Chem. Theory Comput.

Self Cite

View full text Add to dashboard Cite

show abstract

OPUS-Rota3: Improving Protein Side-Chain Modeling by Deep Neural Networks and Ensemble Methods

Gang

Wang

2020

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

Side-chain modeling is critical for protein structure prediction since the uniqueness of the protein structure is largely determined by its side-chain packing conformation. In this paper, differing from most approaches that rely on rotamer library sampling, we first propose a novel sidechain rotamer prediction method based on deep neural networks, named OPUS-RotaNN. Then, on the basis of our previous work OPUS-Rota2, we propose an open-source side-chain modeling framework, OPUS-Rota3, which integrates the results of different methods into its rotamer library as the sampling candidates. By including OPUS-RotaNN into OPUS-Rota3, we conduct our experiments on three native backbone test sets and one non-native backbone test set. On the native backbone test set, CAMEO-Hard61 for example, OPUS-Rota3 successfully predicts 51.14% of all side-chain dihedral angles with a tolerance criterion of 20°and outperforms OSCAR-star (50.87%), SCWRL4 (50.40%), and FASPR (49.85%). On the non-native backbone test set DB379-ITASSER, the accuracy of OPUS-Rota3 is 52.49%, better than OSCAR-star (48.95%), FASPR (48.69%), and SCWRL4 (48.29%). All the source codes including the training codes and the data we used are available at https://github.com/thuxugang/opus_rota3.

show abstract

OPUS-TASS: a protein backbone torsion angles and secondary structure predictor based on ensemble neural networks

2020

View full text Add to dashboard Cite

Motivation Predictions of protein backbone torsion angles (ϕ and ψ) and secondary structure from sequence are crucial subproblems in protein structure prediction. With the development of deep learning approaches, their accuracies have been significantly improved. To capture the long-range interactions, most studies integrate bidirectional recurrent neural networks into their models. In this study, we introduce and modify a recently proposed architecture named Transformer to capture the interactions between the two residues theoretically with arbitrary distance. Moreover, we take advantage of multitask learning to improve the generalization of neural network by introducing related tasks into the training process. Similar to many previous studies, OPUS-TASS uses an ensemble of models and achieves better results. Results OPUS-TASS uses the same training and validation sets as SPOT-1D. We compare the performance of OPUS-TASS and SPOT-1D on TEST2016 (1213 proteins) and TEST2018 (250 proteins) proposed in the SPOT-1D paper, CASP12 (55 proteins), CASP13 (32 proteins) and CASP-FM (56 proteins) proposed in the SAINT paper, and a recently released PDB structure collection from CAMEO (93 proteins) named as CAMEO93. On these six test sets, OPUS-TASS achieves consistent improvements in both backbone torsion angles prediction and secondary structure prediction. On CAMEO93, SPOT-1D achieves the mean absolute errors of 16.89 and 23.02 for ϕ and ψ predictions, respectively, and the accuracies for 3- and 8-state secondary structure predictions are 87.72 and 77.15%, respectively. In comparison, OPUS-TASS achieves 16.56 and 22.56 for ϕ and ψ predictions, and 89.06 and 78.87% for 3- and 8-state secondary structure predictions, respectively. In particular, after using our torsion angles refinement method OPUS-Refine as the post-processing procedure for OPUS-TASS, the mean absolute errors for final ϕ and ψ predictions are further decreased to 16.28 and 21.98, respectively. Availability and implementation The training and the inference codes of OPUS-TASS and its data are available at https://github.com/thuxugang/opus_tass. Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

OPUS-Refine: A Fast Sampling-Based Framework for Refining Protein Backbone Torsion Angles and Global Conformation

Cited by 4 publications

References 32 publications

OPUS-Fold: An Open-Source Protein Folding Framework Based on Torsion-Angle Sampling

OPUS-Fold: An Open-Source Protein Folding Framework Based on Torsion-Angle Sampling

OPUS-Rota3: Improving Protein Side-Chain Modeling by Deep Neural Networks and Ensemble Methods

OPUS-TASS: a protein backbone torsion angles and secondary structure predictor based on ensemble neural networks

Contact Info

Product

Resources

About