Physical-aware model accuracy estimation for protein complex using deep learning method
With the breakthrough of AlphaFold2 on monomers, the research focus of structure prediction has shifted to protein complexes, driving the continued development of new methods for multimer structure prediction. Therefore, it is crucial to accurately estimate quality scores for the multimer model inde...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Computational and Structural Biotechnology Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2001037025000194 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583847859978240 |
|---|---|
| author | Haodong Wang Meng Sun Lei Xie Dong Liu Guijun Zhang |
| author_facet | Haodong Wang Meng Sun Lei Xie Dong Liu Guijun Zhang |
| author_sort | Haodong Wang |
| collection | DOAJ |
| description | With the breakthrough of AlphaFold2 on monomers, the research focus of structure prediction has shifted to protein complexes, driving the continued development of new methods for multimer structure prediction. Therefore, it is crucial to accurately estimate quality scores for the multimer model independent of the used prediction methods. In this work, we propose a physical-aware deep learning method, DeepUMQA-PA, to evaluate the residue-wise quality of protein complex models. Given the input protein complex model, the residue-based contact area and orientation features were first constructed using Voronoi tessellation, representing the potential physical interactions and hydrophobic properties. Then, the relationship between local residues and the overall complex topology as well as the inter-residue evolutionary information are characterized by geometry-based features, protein language model embedding representation, and knowledge-based statistical potential features. Finally, these features are fed into a fused network architecture employing equivalent graph neural network and ResNet network to estimate residue-wise model accuracy. Experimental results on the CASP15 test set demonstrate that our method outperforms the state-of-the-art method DeepUMQA3 by 3.69 % and 3.49 % on Pearson and Spearman, respectively. Notably, our method achieved 16.8 % and 15.5 % improvement in Pearson and Spearman, respectively, for the evaluation of nanobody-antigens. In addition, DeepUMQA-PA achieved better MAE scores than AlphaFold-Multimer and AlphaFold3 self-assessment methods on 43 % and 50 % of the targets, respectively. All these results suggest that physical-aware information based on the area and orientation of atom-atom and atom-solvent contacts has the potential to capture sequence-structure-quality relationships of proteins, especially in the case of flexible proteins. The DeepUMQA-PA server is freely available at http://zhanglab-bioinf.com/DeepUMQA-PA/. |
| format | Article |
| id | doaj-art-6a09f90db526463c8ccd0d7a2f854a1a |
| institution | Kabale University |
| issn | 2001-0370 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Computational and Structural Biotechnology Journal |
| spelling | doaj-art-6a09f90db526463c8ccd0d7a2f854a1a2025-01-28T04:14:34ZengElsevierComputational and Structural Biotechnology Journal2001-03702025-01-0127478487Physical-aware model accuracy estimation for protein complex using deep learning methodHaodong Wang0Meng Sun1Lei Xie2Dong Liu3Guijun Zhang4College of Information Engineering, Zhejiang University of Technology, Hangzhou 310023, ChinaCollege of Information Engineering, Zhejiang University of Technology, Hangzhou 310023, ChinaCollege of Information Engineering, Zhejiang University of Technology, Hangzhou 310023, ChinaCollege of Information Engineering, Zhejiang University of Technology, Hangzhou 310023, ChinaCorresponding author.; College of Information Engineering, Zhejiang University of Technology, Hangzhou 310023, ChinaWith the breakthrough of AlphaFold2 on monomers, the research focus of structure prediction has shifted to protein complexes, driving the continued development of new methods for multimer structure prediction. Therefore, it is crucial to accurately estimate quality scores for the multimer model independent of the used prediction methods. In this work, we propose a physical-aware deep learning method, DeepUMQA-PA, to evaluate the residue-wise quality of protein complex models. Given the input protein complex model, the residue-based contact area and orientation features were first constructed using Voronoi tessellation, representing the potential physical interactions and hydrophobic properties. Then, the relationship between local residues and the overall complex topology as well as the inter-residue evolutionary information are characterized by geometry-based features, protein language model embedding representation, and knowledge-based statistical potential features. Finally, these features are fed into a fused network architecture employing equivalent graph neural network and ResNet network to estimate residue-wise model accuracy. Experimental results on the CASP15 test set demonstrate that our method outperforms the state-of-the-art method DeepUMQA3 by 3.69 % and 3.49 % on Pearson and Spearman, respectively. Notably, our method achieved 16.8 % and 15.5 % improvement in Pearson and Spearman, respectively, for the evaluation of nanobody-antigens. In addition, DeepUMQA-PA achieved better MAE scores than AlphaFold-Multimer and AlphaFold3 self-assessment methods on 43 % and 50 % of the targets, respectively. All these results suggest that physical-aware information based on the area and orientation of atom-atom and atom-solvent contacts has the potential to capture sequence-structure-quality relationships of proteins, especially in the case of flexible proteins. The DeepUMQA-PA server is freely available at http://zhanglab-bioinf.com/DeepUMQA-PA/.http://www.sciencedirect.com/science/article/pii/S2001037025000194Estimation of model accuracySingle-model methodProtein complex structure prediction |
| spellingShingle | Haodong Wang Meng Sun Lei Xie Dong Liu Guijun Zhang Physical-aware model accuracy estimation for protein complex using deep learning method Computational and Structural Biotechnology Journal Estimation of model accuracy Single-model method Protein complex structure prediction |
| title | Physical-aware model accuracy estimation for protein complex using deep learning method |
| title_full | Physical-aware model accuracy estimation for protein complex using deep learning method |
| title_fullStr | Physical-aware model accuracy estimation for protein complex using deep learning method |
| title_full_unstemmed | Physical-aware model accuracy estimation for protein complex using deep learning method |
| title_short | Physical-aware model accuracy estimation for protein complex using deep learning method |
| title_sort | physical aware model accuracy estimation for protein complex using deep learning method |
| topic | Estimation of model accuracy Single-model method Protein complex structure prediction |
| url | http://www.sciencedirect.com/science/article/pii/S2001037025000194 |
| work_keys_str_mv | AT haodongwang physicalawaremodelaccuracyestimationforproteincomplexusingdeeplearningmethod AT mengsun physicalawaremodelaccuracyestimationforproteincomplexusingdeeplearningmethod AT leixie physicalawaremodelaccuracyestimationforproteincomplexusingdeeplearningmethod AT dongliu physicalawaremodelaccuracyestimationforproteincomplexusingdeeplearningmethod AT guijunzhang physicalawaremodelaccuracyestimationforproteincomplexusingdeeplearningmethod |