De novo protein structure prediction by incremental inter-residue geometries prediction and model quality assessment using deep learning

Liu, Jun; He, Guangxing; Zhao, Kailong; Zhang, Guijun

doi:10.1101/2022.01.11.475831

Cited by 12 publications

(11 citation statements)

References 78 publications

(79 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, we used GeomNet, a geometric constraints prediction network in our recently developed structure prediction server RocketX (Liu et al ., 2022), to predict the inter-residue distance of full-chain. For the query sequence, MSA was generated by iterative search against UniRef30 (Mirdita et al ., 2017) and BFD (Steinegger et al ., 2019) databases by using HHblits (Steinegger et al ., 2019) with gradually relaxed e-values of 1e -30 , 1e -10 , 1e -6 , and 1e -3 .…”

Section: Methodsmentioning

confidence: 99%

Structural analogue-based protein structure domain assembly assisted by deep learning

Peng

Zhou

Xia

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Motivation: With the breakthrough of AlphaFold2, the protein structure prediction problem has made a remarkable progress through end-to-end deep learning techniques, in which correct folds could be built for nearly all single-domain proteins. However, the full-chain modelling appears to be lower on average accuracy than that for the constituent domains and requires higher demand on computing hardware, indicating the performance of full-chain modelling still needs to be improved. In this study, we investigate whether the predicted accuracy of full-chain model can be further improved by domain assembly assisted by deep learning. Results: In this article, we developed a structural analogue-based protein structure domain assembly method assisted by deep learning, named SADA. In SADA, a multi-domain protein structure database (MPDB) was constructed for the full-chain analogue detection using individual domain models. Starting from the initial model constructed from the analogue, the domain assembly simulation was performed to generate the full-chain model through a two-stage differential evolution algorithm guided by the physics-based force field and an inter-residue distance potential predicted by deep learning. SADA was compared with the state-of-the-art domain assembly methods on 356 benchmark proteins, and the average TM-score of SADA models is 8.1% and 27.0% higher than that of DEMO and AIDA, respectively. We also assembled full-chain models of 20 human multi-domain proteins using individual domain models independently predicted by AlphaFold2, where the SADA full-chain models obtained a 4.8% higher average TM-score than full-chain models directly predicted by AlphaFold2 and fewer computing resources were required. In addition, we also find that the domains often interact in the similar way in the quaternary orientations if the domains have similar tertiary structures. Furthermore, homologous templates and structural analogues are complementary for multi-domain protein full-chain modelling.

show abstract

Section: Methodsmentioning

confidence: 99%

Structural analogue-based protein structure domain assembly assisted by deep learning

Peng

Zhou

Xia

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this study, we used GeomNet, a geometric constraints prediction network in our recently developed structure prediction server RocketX (Liu et al, 2022), to predict the inter-residue distance of full-chain. For the query sequence, MSA was generated by iterative search against UniRef30 (Mirdita et al, 2017) and BFD databases by using HHblits with gradually relaxed e-values of 1e -30 , 1e -10 , 1e -6 , and 1e -3 .…”

Section: Force Field For Domain Assemblymentioning

confidence: 99%

“…With the development of deep learning, protein structure prediction methods have been developed successively, such as trRosetta (Yang et al, 2020;Su et al, 2021;Du et al, 2021), RaptorX (Xu, 2019;Xu and Wang, 2019), RocketX (Liu et al, 2022), and D-I-TASSER (Zheng et al, 2021). Recently, end-to-end methods, such as AlphaFold2 (Jumper et al, 2021) and RoseTTAFold (Baek et al, 2021), have been proposed to accurately predict more complex proteins including some multi-domain proteins.…”

Section: Introductionmentioning

confidence: 99%

Structural analogue-based protein structure domain assembly assisted by deep learning

et al. 2022

View full text Add to dashboard Cite

Motivation With the breakthrough of AlphaFold2, the protein structure prediction problem has made remarkable progress through deep learning end-to-end techniques, in which correct folds could be built for nearly all single-domain proteins. However, the full-chain modelling appears to be lower on average accuracy than that for the constituent domains and requires higher demand on computing hardware, indicating the performance of full-chain modelling still needs to be improved. In this study, we investigate whether the predicted accuracy of the full-chain model can be further improved by domain assembly assisted by deep learning. Results In this article, we developed a structural analogue-based protein structure domain assembly method assisted by deep learning, named SADA. In SADA, a multi-domain protein structure database (MPDB) was constructed for the full-chain analogue detection using individual domain models. Starting from the initial model constructed from the analogue, the domain assembly simulation was performed to generate the full-chain model through a two-stage differential evolution algorithm guided by the energy function with an inter-residue distance potential predicted by deep learning. SADA was compared with the state-of-the-art domain assembly methods on 356 benchmark proteins, and the average TM-score of SADA models is 8.1% and 27.0% higher than that of DEMO and AIDA, respectively. We also assembled 293 human multi-domain proteins, where the average TM-score of the full-chain model after the assembly by SADA is 1.1% higher than that of the model by AlphaFold2. To conclude, we find that the domains often interact in the similar way in the quaternary orientations if the domains have similar tertiary structures. Furthermore, homologous templates and structural analogues are complementary for multi-domain protein full-chain modelling. Availability http://zhanglab-bioinf.com/SADA Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

“…In our previously proposed SADA approach, a multi-domain protein structure database (MPDB) was constructed for the full-chain analogue detection using individual domain models. Based on the detected analogue, an energy function assisted by deep learning network GeomNet is designed to guide domain assembly [25]. The results show that SADA can be an effective complement to AlphaFold2 in multi-domain protein modelling [21].…”

Section: Introductionmentioning

confidence: 99%

“…Obviously, for 3dom and m4dom, SADA has lower accuracy, because it is difficult to detect effective structural analogues when the number of structural domains increases. However, the DeepIDDP can provide more inter-domain information than distance prediction methods (such as GeomNet) to capture the orientation between domains, and further improve the accuracy of the final model [25]. ≥ 0.5), representing 95.2% of the total and 3.39% higher than SADA.…”

mentioning

confidence: 99%

Inter-domain distance prediction based on deep learning for domain assembly

Zhang

Peng

et al. 2022

Preprint

View full text Add to dashboard Cite

AlphaFold2 achieved a breakthrough in protein structure prediction through the end-to-end deep learning method, which can predict nearly all single-domain proteins at experimental resolution. However, the prediction accuracy of full-chain proteins is generally lower than that of single-domain proteins because of the incorrect interactions between domains. In this work, we develop an inter-domain distance prediction method, named DeepIDDP. In DeepIDDP, we design a neural network with attention mechanisms, where two new inter-domain features are used to enhance the ability to capture the interactions between domains. Furthermore, we propose a data enhancement strategy termed DPMSA, which is employed to deal with the absence of co-evolutionary information on targets. We integrate DeepIDDP into our previously developed domain assembly method SADA, termed SADA-DeepIDDP. Tested on a given multi-domain benchmark dataset, the accuracy of SADA-DeepIDDP inter-domain distance prediction is 11.3% and 21.6% higher than trRosettaX and trRosetta, respectively. The accuracy of the domain assembly model is 2.5% higher than that of SADA. Meanwhile, we reassemble 68 human multi-domain protein models with TM-score less than or equal to 0.80 from the AlphaFold protein structure database, where the average TM-score is improved by 11.8% after the reassembly by our method. The online server is at http://zhanglab-bioinf.com/DeepIDDP/.

show abstract

De novo protein structure prediction by incremental inter-residue geometries prediction and model quality assessment using deep learning

Cited by 12 publications

References 78 publications

Structural analogue-based protein structure domain assembly assisted by deep learning

Structural analogue-based protein structure domain assembly assisted by deep learning

Structural analogue-based protein structure domain assembly assisted by deep learning

Inter-domain distance prediction based on deep learning for domain assembly

Contact Info

Product

Resources

About