Introduction

DeepUMQA-X is a fully automated server for accurate protein single-chain and complex model quality assessment. It evaluates the local residue accuracy (lDDT), overall interface accuracy (QS-score, only complex), and overall folding accuracy (TM-score) of protein structures, based on our-house improved single-model methods with lightweight structure align. DeepUMQA-X is one of the best performing methods, which performs well in CASP16 EMA blind test. In addition, in the CAMEO-QE blind test, the key component of the DeepUMQA-X server, GraphCPLMQA2L, ranked first in the 1-year (2023-06-09 to 2024-06-01) model quality assessment. (More about DeepUMQA-X...)

News

• 2025/04/23, DeepUMQA-X was accepted for publication by Nucleic Acids Research.
• 2024/11/30, Breaking news Ranking of DeepUMQA-X (Group name: GuijunLab-QA & GuijunLab-Human) in blind test of CASP16 EMA:
  • Local accuracy estimation: Third in complex local interface accuracy estimation; Third in complex local interface residue identification.
  • Global accuracy estimation: Third in overall interface accuracy estimation (excluding AC); Fifth in overall fold accuracy estimation (excluding AC).
  • Model selection: Third in homologous complex model selection; Fourth in heterology complex model selection.
  • Self-Assessment: Second in per-residue accuracy assessment; Fourth in per-atom accuracy assessment.
• 2024/06/01, GraphCPLMQA ranked first in the 1-year (2023-06-09 to 2024-06-01) blind test of CAMEO-QE.
• 2023/11/28, GraphCPLMQA was accepted for publication by Briefings in Bioinformatics.
• 2023/9/22, DeepUMQA3 was accepted for publication by Bioinformatics..
• 2022/12/13, Breaking news DeepUMQA3 (Group name: GuijunLab-RocketX) ranked first in the complex interface residue accuracy estimation of CASP15.
• 2022/12/2, DeepUMQA2 was accepted for publication by Briefings in Bioinformatics..
• 2022/11/26, DeepUMQA ranked first in the 1-year (2021-12-03 to 2022-11-26) blind test of CAMEO-QE.
• 2022/02/03, DeepUMQA was accepted for publication by Bioinformatics.
• 2021/12/22, DeepUMQA ranked first in the 1-month (2021-11-26 to 2021-12-18) blind test of CAMEO-QE.

References

• Dong Liu, Jun Liu, Haodong Wang, Fang Liang, Guijun Zhang*. DeepUMQA-X: Comprehensive and insightful estimation of model accuracy for protein single-chain and complex.   Nucleic Acids Research, 2025, https://doi.org/10.1093/nar/gkaf380.
• Dong Liu^✝, Biao Zhang^✝, Jun Liu, Hui Li*, Le Song* and Guijun Zhang*. GraphCPLMQA: Assessing protein model quality based on deep graph coupled networks using protein language model.   Briefings in Bioinformatics, 2023, 25(1):bbad420.
• Jun Liu, Dong Liu, Guijun Zhang*. DeepUMQA3: a web server for accurate assessment of interface residue accuracy in protein complexes.   Bioinformatics, 2023, 39(10): btad591.
• Jun Liu, Dong Liu, Guangxing He, Guijun Zhang*. Estimating protein complex model accuracy based on ultrafast shape recognition and deep learning in CASP15.   Proteins: Structure, Function, and Bioinformatics, 2023, 91(12): 1861-1870.
• Jun Liu, Kailong Zhao and Guijun Zhang*. Improved model quality assessment using sequence and structural information by enhanced deep neural networks.   Briefings in Bioinformatics, 2023, 24(1): bbac507.
• Saisai Guo^✝, Jun Liu^✝, Xiaogen Zhou, Guijun Zhang*. DeepUMQA: Ultrafast Shape Recognition-based Protein Model Quality Assessment using Deep Learning.   Bioinformatics, 2022, 38(7): 1895-1903.