DockQ: A Quality Measure for Protein-Protein Docking Models
Open accessThe DOI matches its Crossref record; 26 of the 26 references it cites were found in a registry or an index.
Abstract
The state-of-the-art to assess the structural quality of docking models is currently based on three related yet independent quality measures: Fnat, LRMS, and iRMS as proposed and standardized by CAPRI. These quality measures quantify different aspects of the quality of a particular docking model and need to be viewed together to reveal the true quality, e.g. a model with relatively poor LRMS (>10Å) might still qualify as 'acceptable' with a descent Fnat (>0.50) and iRMS (<3.0Å). This is also the reason why the so called CAPRI criteria for assessing the quality of docking models is defined by applying various ad-hoc cutoffs on these measures to classify a docking model into the four classes: Incorrect, Acceptable, Medium, or High quality. This classification has been useful in CAPRI, but since models are grouped in only four bins it is also rather limiting, making it difficult to rank models, correlate with scoring functions or use it as target function in machine learning algorithms. Here, we present DockQ, a continuous protein-protein docking model quality measure derived by combining Fnat, LRMS, and iRMS to a single score in the range [0, 1] that can be used to assess the quality of protein docking models. By using DockQ on CAPRI models it is possible to almost completely reproduce the original CAPRI classification into Incorrect, Acceptable, Medium and High quality. An average PPV of 94% at 90% Recall demonstrating that there is no need to apply predefined ad-hoc cutoffs to classify docking models. Since DockQ recapitulates the CAPRI classification almost perfectly, it can be viewed as a higher resolution version of the CAPRI classification, making it possible to estimate model quality in a more quantitative way using Z-scores or sum of top ranked models, which has been so valuable for the CASP community. The possibility to directly correlate a quality measure to a scoring function has been crucial for the development of scoring functions for protein structure prediction, and DockQ should be useful in a similar development in the protein docking field. DockQ is available at http://github.com/bjornwallner/DockQ/.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures



Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“In a comparison with AlphaFold-Multimer (45) on a dataset of 2,978 recent multimeric complexes deposited in the PDB, ESMFold achieves the same qualitative DockQ (39) categorization for 53.2% of chain pairs, despite not being trained on protein complexes (Fig. S6).”
From Evolutionary-scale prediction of atomic level protein structure with a language model (2022) · cites this paper as [39]
The cited source says what the claim attributes to it.
In the citing paper
“Generally, protein–protein prediction success (DockQ > 0.23)40 has increased (paired Wilcoxon signed-rank test, P = 1.8 × 10−18)”
From Accurate structure prediction of biomolecular interactions with AlphaFold 3 (2024) · cites this paper as [40]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“As the bound form of the proteins is used, this should represent an easy case for GRAMM-based docking, and performance drops significantly when unbound structures or models are used53.”
From Improved prediction of protein-protein interactions using AlphaFold2 (2021) · cites this paper as [53]
Experimental — accuracy not yet measuredIn the citing paper
“the overall top models from AF2Complex are acceptable or better in 15 of 17 (88%) cases, and 13 (76%) models are of medium or high quality according to the DockQ score42 (Fig. 3a, Supplementary Data 1), with a mean of 0.62”
From AF2Complex predicts direct physical interactions in multimeric proteins with deep learning (2022) · cites this paper as [42]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Scoring function for automated assessment of protein structure template quality
Zhang, Y., & Skolnick, J. (2004) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.20264 · Cited by 2,392
ZDOCK: An initial‐stage protein‐docking algorithm
Chen, R., Li, L., & Weng, Z. (2003) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.10389 · Cited by 1,393 · cited 1×
“Thus, the scoring part of CAPRI [8] and benchmarks of scoring functions for docking [11,10,12,13] only focuses on the ability to select good models according to the CAPRI classification, completely ignoring the potential useful information in the lower ranked models in assessing the ability to…”
MethodAssessment of blind predictions of protein–protein interactions: Current status of docking methods
Méndez, R., Leplae, R., De Maria, L., & Wodak, S. J. (2003) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.10393 · Cited by 418 · cited 2×
“The backbone atoms of these 'interface' residues is then superposed on their equivalents in the predicted complex (model) to compute the iRMS [9].”
BackgroundMethodMaxSub: an automated measure for the assessment of protein structure prediction quality
Siew, N., Elofsson, A., Rychlewski, L., & Fischer, D. (2000) · Bioinformatics · doi:10.1093/bioinformatics/16.9.776 · Cited by 401 · cited 1×
“In contrast to the protein structure prediction field, where there are several widely accepted quality measures; e.g., C α -RMSD, GDT_TS [3], MaxSub [4], TM-score [5], and S-score [6], the IS-score [7] for assessing protein complex models has not achieved wide adoption by the field and the current…”
ResultDocking and scoring protein complexes: CAPRI 3rd Edition
Lensink, M. F., Méndez, R., & Wodak, S. J. (2007) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.21804 · Cited by 340 · cited 4×
“…standard, the average iRMS for the models with Medium quality classified incorrectly as High by DockQ is 1.05Å, and High quality classified incorrectly as Medium is 0.93Å, while the cutoff in iRMS between medium and high quality is 1.0Å (Medium < 1.0Å; High (cid:1) 1.0Å) according to CAPRI [10].”
BackgroundMethodInfluentialStructure of the Cytoplasmic β Subunit--T1 Assembly of Voltage-Dependent K + Channels
Gulbis, J. M., Zhou, M., Mann, S., & MacKinnon, R. (2000) · Science · doi:10.1126/science.289.5476.123 · Cited by 320
3did: a catalog of domain-based interactions of known three-dimensional structure
Mosca, R., Céol, A., Stein, A., Olivella, R., & Aloy, P. (2014) · Nucleic Acids Research · doi:10.1093/nar/gkt887 · Cited by 305 · cited 1×
“…in traditional structural biology and the structural genomics projects that aim at high-throughput complex structure determination [1], the latest statistics from 3did database [2] show that only 7% of the known protein interactions in humans have an associated experimental complex structure.”
BackgroundSwarmDock: a server for flexible protein–protein docking
Torchala, M., Moal, I. H., Chaleil, R. A. G., Fernandez-Recio, J., & Bates, P. A. (2013) · Bioinformatics · doi:10.1093/bioinformatics/btt038 · Cited by 285 · cited 1×
“This set contained 56,015 docking models for 118 targets from the protein-protein docking Benchmark 4.0 [17], constructed using SwarmDock [18] gra-ciously provided by the authors of Moal et al [13].”
MethodInfluentialProcessing and analysis of CASP3 protein structure predictions
Zemla, A., Venclovas, ?., Moult, J., & Fidelis, K. (1999) · Proteins: Structure, Function, and Genetics · doi:10.1002/(sici)1097-0134(1999)37:3+<22::aid-prot5>3.0.co;2-w · Cited by 282 · cited 1×
“In contrast to the protein structure prediction field, where there are several widely accepted quality measures; e.g., C α -RMSD, GDT_TS [3], MaxSub [4], TM-score [5], and S-score [6], the IS-score [7] for assessing protein complex models has not achieved wide adoption by the field and the current…”
ResultDocking, scoring, and affinity prediction in CAPRI
Lensink, M. F., & Wodak, S. J. (2013) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.24428 · Cited by 249
Cited by
Items 201–300 of 475
Mitonuclear Sex Determination? Empirical Evidence from Bivalves
Smith, C. H., Mejia-Trujillo, R., Breton, S., Pinto, B. J., Kirkpatrick, M., & Havird, J. C. (2023) · Molecular Biology and Evolution · doi:10.1093/molbev/msad240 · Cited by 10
“Mitonuclear Sex in Bivalves · https://doi.org/10.1093/molbev/msad240 MBE no acceptable support for PPIs between the GCNT1 protein and any of the mitochondrial proteins investigated, with the strongest interaction being between F-ORF and GCNT1 (supplementary fig.”
BackgroundInfluentialÜnsal, S., Holland, B., Sardag, I., & Timuçin, E. (2026) · Briefings in Bioinformatics · doi:10.1093/bib/bbag137 · Cited by 11
Feldman, J., Brogi, M., & Skolnick, J. (2026) · Computational and Structural Biotechnology Journal · doi:10.34133/csbj.0142 · Cited by 14
“For multimeric proteins with more than two chains, the weighted average DockQ score across all pairwise chain interfaces was computed and used as the reported metric, with a standard threshold of 0.23 used for a valid interfacial conformation [44, 59].”
InfluentialMayol, G. F., Defelipe, L. A., Arcon, J. P., Turjanski, A. G., & Marti, M. A. (2022) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.2c00264 · Cited by 10
Massive Sampling Strategy for Antibody–Antigen Targets in CAPRI Round 55 With MassiveFold
Raouraoua, N., Lensink, M. F., & Brysbaert, G. (2025) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.26802 · Cited by 10
Prediction of CD28-CD86 protein complex structure using different level of resolution approach
Krupa, P., Spodzieja, M., & Sieradzan, A. K. (2020) · Journal of Molecular Graphics and Modelling · doi:10.1016/j.jmgm.2020.107802 · Cited by 10
Strengths and limitations of web servers for the modeling of TCRpMHC complexes
Le, H. N., de Freitas, M. V., & Antunes, D. A. (2024) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2024.06.028 · Cited by 10
“In order to better characterize the contribution of inaccurate TCR docking orientation to our “all protein” RMSD results, we conducted a series of additional evaluations using (i) the CAPRI assessment [34], (ii) the DockQ score [41], and (iii) the measurement of the TCR angles with TCR3d [36].”
MethodExploring the potential of structure-based deep learning approaches for T cell receptor design
Filho, H. V. R., Jara, G. E., Guerra, J. V. D. S., Cheung, M., Felbinger, N. R., Pereira, J. G. D. C., Pierce, B. G., & de Oliveira, P. S. L. (2024) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1012489 · Cited by 9
Human genetic variants in SLC39A8 impact uptake and steady-state metal levels within the cell
Wang, W., Garofoli, A., Ferrada, E., Klimek, C., Steurer, B., Inglés‐Prieto, Á., Osthushenrich, T., MacNamara, A., Mälarstig, A., Wiedmer, T., et al. (2025) · Life Science Alliance · doi:10.26508/lsa.202403028 · Cited by 11
“In addition, we used a protein – protein interaction score (pDockQ score) that accounts for the quality of the complex (48, 49).”
Han, B., Zhang, Y., Li, L., Gong, X., & Xia, K. (2025) · Briefings in Bioinformatics · doi:10.1093/bib/bbaf083 · Cited by 10
“The pairwise comparison is measured by scores like QS [21, 22], lDDT [23], or DockQ [24], (with ModFOLDdock and MULTICOM_qa as notable examples [20, 25]), the average of the scores is used to assess the quality of this structure.”
Albuquerque, A., da Silva, H. C., Sartori, G., & da Silva, J. H. M. (2021) · Journal of Biomolecular Structure and Dynamics · doi:10.1080/07391102.2021.1885492 · Cited by 9
“The DockQ program segregated all docked complexes into four CAPRI quality categories, comparing them with the original crystal structure, and identifying the models closest to native (Basu & Wallner, 2016).”
MethodShuvo, M. H., Karim, M., Roche, R., & Bhattacharya, D. (2023) · Bioinformatics Advances · doi:10.1093/bioadv/vbad070 · Cited by 11
“…as follows: where d i represents the scaling factors: d 1 for LRSM and d 2 for iRMS. d 1 and d 2 are optimized to be 8.5 and 1.5 A˚, respectively, based on the ability to separate models according to CAPRI classifications in terms of F 1 scores (Basu and Wallner, 2016; Lensink and Wodak, 2013).”
MethodInfluentialOuter membrane β‐barrel structure prediction through the lens of AlphaFold2
Topitsch, A., Schwede, T., & Pereira, J. (2024) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.26552 · Cited by 9
Modeling membrane geometries implicitly in Rosetta
Woods, H., Leman, J. K., & Meiler, J. (2024) · Protein Science · doi:10.1002/pro.4908 · Cited by 9
Challenges and compromises: Predicting unbound antibody structures with deep learning
Greenshields‐Watson, A., Vavourakis, O., Spoendlin, F. C., Cagiada, M., & Deane, C. M. (2025) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2025.102983 · Cited by 12
Improved cluster ranking in protein–protein docking using a regression approach
Sotudian, S., Desta, I., Hashemi, N., Zarbafian, S., Kozakov, D., Vakili, P., Vajda, S., & Paschalidis, I. C. (2021) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2021.04.028 · Cited by 10
“[29] verified that DockQ matched with previous CAPRI experiments well.”
BackgroundResultInfluentialScoring of protein–protein docking models utilizing predicted interface residues
Pozzati, G., Kundrotas, P. J., & Elofsson, A. (2022) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.26330 · Cited by 10
Atomic-level evolutionary information improves protein–protein interface scoring
Quignot, C., Granger, P., Chacón, P., Guérois, R., & Andréani, J. (2021) · Bioinformatics · doi:10.1093/bioinformatics/btab254 · Cited by 9
The pdb2sql Python Package: Parsing, Manipulation and Analysis of PDB Files Using SQL Queries
Renaud, N., & Geng, C. (2020) · The Journal of Open Source Software · doi:10.21105/joss.02077 · Cited by 9
“The following snippet shows how to compute these measures: from pdb2sql import StructureSimilarity sim = StructureSimilarity(decoy = '1AK4_model.pdb', ref = '1AK4_xray.pdb') irmsd = sim.compute_irmsd_fast() lrmsd = sim.compute_lrmsd_fast() fnat = sim.compute_fnat_fast() dockQ = sim.compute_DockQScore(fnat, lrmsd, irms…”
MethodInfluentialTokenizing Foldable Protein Structures with Machine-Learned Artificial Amino-Acid VocabularyPreprint
Lin, X., Chen, Z., Li, Y., Ma, Z., Fan, C., Cao, Z., Feng, S., Gao, Y. Q., & Zhang, J. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.11.27.568722 · Cited by 9
“Figure 4c shows that the interface is relatively accurate(Basu and Wallner, 2016), though not perfect, with an average and median DockQ exceeding 0.49.”
BackgroundMethodInfluentialZhang, Z., Jin, R., Liu, J., Xu, G., Wang, X., Žitnik, M., Zhao, H., Cong, L., & Wang, M. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.10.23.619960 · Cited by 9
“We measure the quality of protein-protein interfaces using DockQ [ 72 ].”
Baptista, D., Gomez-Lucas, L., Jänes, J., Krogan, N. J., Amorim, M. J., Ivarsson, Y., & Beltrão, P. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.06.04.657796 · Cited by 12
“…the corresponding experimental structures and scoring them using two structural similarity metrics, the TM-score (Zhang & Skolnick, 2004), a measure of the global similarity between two structures, and the DockQ score (Basu & Wallner, 2016), which measures the quality of the interaction interface.”
End-to-end cryo-EM complex structure determination with high accuracy and ultra-fast speed
Wang, J., Tan, C., Gao, Z., Zhang, G., Zhang, Y., & Li, S. Z. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-025-01056-0 · Cited by 8
Evaluation of AlphaFold modeling for elucidation of nanobody–peptide epitope interactions
Sachdev, S., Roy, S., Saha, S. J., Zhao, G., Kumariya, R., Creemer, B. A., Yin, R., Pierce, B. G., Bewley, C. A., & Cheloha, R. W. (2025) · Journal of Biological Chemistry · doi:10.1016/j.jbc.2025.110268 · Cited by 8
“Model accuracy metrics were computed based on comparison of modeled complexes with the corresponding X-ray structures using the DockQ program (54).”
Xu, Q., Liu, W., Liu, H., & Zheng, Z. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c00387 · Cited by 11
AI-first structural identification of pathogenic protein target interfaces
Saluri, M., Landreh, M., & Bryant, P. (2025) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1013168 · Cited by 8
“The reason for using the TM-score instead of the DockQ [13] score is because the full genetic sequences were used here, meaning that the predicted and native structures have different lengths, something DockQ handles poorly (see Fig B in S1 Appendix).”
Kagaya, Y., Nakamura, T., Verburgt, J. C., Jain, A., Terashi, G., Punuru, P. D., Tugolukova, E., Park, J. H., Saha, A., Huang, D. Y. C., et al. (2025) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.70033 · Cited by 12
“…starts by MSA construction using rMSA and metagenome databases, structure modeling using seven methods, followed by initial scoring using ranksum, and final manual model selection incorporating expert review and curation. scoring interface such as QS-score [78] and DockQ [79] have been employed.”
State-Specific Peptide Design Targeting G Protein-Coupled Receptors
Yang, X. W., Wang, H., Li, J., Hu, J., Chen, Z., Zheng, Z., Liu, L., Zhu, K., He, J., Gong, H., et al. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c00884 · Cited by 9
Mahtha, S. K., Sureshkumar, V., & Mohanty, D. (2026) · NAR Genomics and Bioinformatics · doi:10.1093/nargab/lqag002 · Cited by 12
“DockQ integrates several interface-specific measures, including interface RMSD (iRMSD), ligand RMSD (LRMSD), and the fraction of native contacts pre-served (Fnat) [ 39 ].”
Guidelines To Predict Binding Poses of Antibody–Integrin Complexes
Chaves, B., Sartori, G., VASCONCELOS, D. B., Savino, W., Caffarena, E. R., Cotta-de-Almeida, V., & da Silva, J. H. M. (2020) · ACS Omega · doi:10.1021/acsomega.0c00226 · Cited by 10
FTDMP : A Framework for Protein–Protein, Protein– DNA, and Protein– RNA Docking and Scoring
Olechnovič, K., Banciul, R., Dapkūnas, J., & Venclovas, Č. (2025) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.26792 · Cited by 9
Learning with Privileged Knowledge Distillation for Improved Peptide–Protein Docking
Zhang, Z., Verburgt, J. C., Kagaya, Y., Christoffer, C. W., & Kihara, D. (2025) · ACS Omega · doi:10.1021/acsomega.5c00967 · Cited by 11
Zhang, W., Meng, Q., Tang, J., & Guo, F. (2021) · Briefings in Bioinformatics · doi:10.1093/bib/bbab150 · Cited by 9
PointDE: Protein Docking Evaluation Using 3D Point Cloud Neural Network
Chen, Z. X., Liu, N., Huang, Y., Min, X., Zeng, X., Ge, S., Zhang, J., & Xia, N. (2023) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2023.3279019 · Cited by 8
“We calculated these three parameters through DockQ program [41].”
MethodIterative In Silico Screening for Optimizing Stable Conformation of Anti-SARS-CoV-2 Nanobodies
Shang, W., Hu, X., Lin, X., Li, S., Xiong, S., Huang, B., & Wang, X. (2024) · Pharmaceuticals · doi:10.3390/ph17040424 · Cited by 8
“For each antigen–antibody complex, ClusPro retained 1000 lowest energy models and evaluated the top 30 models (centers of the 30 largest clusters) using the DockQ program [43].”
MethodMapping the FtsQBL divisome components in bacterial NTD pathogens as potential drug targets
Kaur, H., & Lynn, A. M. (2023) · Frontiers in Genetics · doi:10.3389/fgene.2022.1010870 · Cited by 10
Reliable protein-protein docking with AlphaFold, Rosetta, and replica-exchangePreprint
Harmalkar, A., Lyskov, S., & Gray, J. J. (2024) · eLife · doi:10.7554/elife.94029.1 · Cited by 12
(2026) · doi:10.7554/elife.98179.2 · Cited by 8
“…sought to use only those models that were correctly predicted by AF2: we therefore filtered the models and retained only those with a DockQ score ( Basu and Wallner, 2016 ) of at least 0.23 relative to the corresponding experimentally defined structure (0.23 is the quality cutoff defined to be an…”
InfluentialFrom Dataset Curation to Unified Evaluation: Revisiting Structure Prediction Benchmarks with PXMeterPreprint
Ma, W., Liu, Z., Yang, J., Lu, C., Zhang, H., & Xiao, W. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.07.17.664878 · Cited by 12
“Most existing evaluation tools [2–5] either provide only a subset of desired metrics, lack robust support for the outputs produced by modern AI models, or are difficult to extend to incorporate additional models or evaluation metrics.”
In silico discovery of nanobody binders to a G-protein coupled receptor using AlphaFold-Multimer
Harvey, E. P., Smith, J. S., Hurley, J. D., Granados, A. J., Schmid, E. W., Liang-Lin, J. G., Zhang, H., Meara, E. M., Wren, E. K., Paul, S., et al. (2026) · Nature Communications · doi:10.1038/s41467-026-72093-5 · Cited by 7
“Among these were average interface predicted aligned error (PAE), average interface predicted local distance difference test (pLDDT), pDockQ 18 ,and average model support (the average number of models in which interface contacts are observed) 16,18 .”
Unified Sampling and Ranking for Protein Docking With DFMDock
Chu, L., Sarma, S., Xu, D., & Gray, J. J. (2026) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.70160 · Cited by 7
Multimodal SARS-CoV-2 interactome sketches the virus-host spatial organization
Dugied, G., Laurent, E. M. N., Attia, M., Gimeno, J., Bachiri, K., Samavarchi‐Tehrani, P., Donati, F., Rahou, Y., Munier, S., Amara, F., et al. (2025) · Communications Biology · doi:10.1038/s42003-025-07933-z · Cited by 7
“Both the number of predicted structures and the pDockQ values are unsurpris-ingly lower than the ones previously reported for human datasets 36 as most of the viral proteins are known to be challenging to predict due to a lack of heterologs in the AlphaFold database 37 .”
InfluentialPrix Nobel de chimie 2024: David Baker, Demis Hassabis et John M. Jumper
Aguero, S., Bettler, E., & Gouet, P. (2025) · médecine/sciences · doi:10.1051/medsci/2025060 · Cited by 9
Wang, Y. X., Wang, F., Feng, L., Zhang, C., & Lai, L. (2025) · Protein Science · doi:10.1002/pro.70338 · Cited by 9
Thianthrenium Chemistry for Identification of Protein–Protein Interactions in Cells
Bohdan, K., Hartmann, P., Müller, S., Marchionni, D., Preisinger, C., Jacobs, J. B., Vogelsang, L., Sterling, M. S., Dietz, K., & Ritter, T. (2025) · Journal of the American Chemical Society · doi:10.1021/jacs.5c16665 · Cited by 10
Machine learning, docking, or physics for structure prediction of ligand-induced ternary complexes
Solazzo, R., Chen, S., & Riniker, S. (2026) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2025.103217 · Cited by 8
Yang, Q., Jin, X., Zhou, H., Ying, J., Zou, J., Liao, Y., Lu, X., Ge, S., Yu, H. J., & Min, X. (2024) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2024.108067 · Cited by 8
Forecasting SARS-CoV-2 spike protein evolution from small data by deep learning and regression
King, S. B., Chen, X. E., Ng, S. W. S., Rostin, K., Hahn, S. V., Roberts, T., Schwab, J. C., Sekhon, P., Kagieva, M., Reilly, T., et al. (2024) · Frontiers in Systems Biology · doi:10.3389/fsysb.2024.1284668 · Cited by 8
Antal, Z., Szövérfi, J., & Fejer, S. N. (2017) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.7b00078 · Cited by 7
Current Computational Methods for Protein-peptide Complex Structure Prediction
Yang, C., Xu, X., & Xiang, C. (2023) · Current Medicinal Chemistry · doi:10.2174/0109298673263447230920151524 · Cited by 7
Soft regions of protein surface are potent for stable dimer formation
Hadi‐Alijanvand, H. (2019) · Journal of Biomolecular Structure and Dynamics · doi:10.1080/07391102.2019.1662328 · Cited by 7
“The thermodynamic stabilities of docking results are computed by utilizing PISA and their similarity to the native structures are measured using DOCKQ (Basu & Wallner, 2016) and they are presented in Figure 6.”
MethodWhen Alphafold2 predictions go wrong for protein–protein complexes, is there something to be learnt?
Martin, J. (2022) · Quarterly Reviews of Biophysics · doi:10.1017/s0033583522000051 · Cited by 7
“Out of these 17 complexes, Alphafold2 multimer achieved correct predictions for 14 cases, as assessed by the DockQ score (Basu and Wallner, 2016).”
PPIscreenML: Structure-based screening for protein-protein interactions using AlphaFoldPreprint
Mischley, V., Maier, J., Chen, J., & Karanicolas, J. (2024) · eLife · doi:10.7554/elife.98179.1 · Cited by 10
Origin-1: a generative AI platform for de novo antibody design against novel epitopesPreprint
Levine, S., King, J. E., Stern, J., Grayson, D., Wang, R., Yin, R., Lupo, U., Kulytė, P., Brand, R. M., Bertin, T., et al. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.01.14.699389 · Cited by 12
Boyd, L. F., Jiang, J., Ahmad, J., Natarajan, K., & Margulies, D. H. (2024) · The Journal of Immunology · doi:10.4049/jimmunol.2300839 · Cited by 6
Chiesa, L., Khasanova, D., & Kellenberger, E. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c00489 · Cited by 7
Nanodesigner: resolving the complex-CDR interdependency with iterative refinement
Zertuche, M. M. R., Kafkas, Ş., Renn, D., Rueping, M., & Hoehndorf, R. (2025) · Journal of Cheminformatics · doi:10.1186/s13321-025-01069-2 · Cited by 7
“We also use DockQ [43] to evaluate docking quality; DockQ combines scores such as interfacial contact preservation, ligand RMSD, and interface RMSD. DockQ is sensitive to the precise orientation and positioning of the antibody relative to the epitope.”
Stratiichuk, R., Kyrylenko, R., Koleiev, I., Savchenko, I., Voitsitskyi, T., Husak, V., Yesylevskyy, S., Starosyla, S., & Nafiiev, A. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c01765 · Cited by 6
AlphaFold3 prediction of protein-protein complex: Is it ready for thermodynamic analysis?
Liu, X., Pang, K., Wu, H., Chong, B., Yin, Z., Sun, Z., & Wang, X. (2025) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2025.10.015 · Cited by 7
“…benchmarks of its predictive power for protein-protein complexes reported in the literature, most benchmarks of AF3 [27] focus on static structures [28] and use metrics such as root-mean-square deviation (RMSD) and DockQ [29] as quantitative evaluators, which do not provide the whole picture.”
Richard, A. C., & Pantazes, R. J. (2024) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.26773 · Cited by 5
“Predicted complexes with an antigen RMSD > 10 Å were defined as poorly docked, a threshold and classification based on the DockQ score used in the Critical Assessment of PRediction of Interactions (CAPRI) [76].”
From Traditional Methods to Deep Learning Approaches: Advances in Protein–Protein Docking
Jiang, L., Zhang, K., Zhu, K., Zhang, H., Shen, C., & Hou, T. (2025) · Wiley Interdisciplinary Reviews Computational Molecular Science · doi:10.1002/wcms.70016 · Cited by 8
“…to design a single robust continuous quality estimate method to cover all the diverse structural attributes individually assessed by the CAPRI measures, thus deriving a variety of indicators, such as Cα-RMSD, TM-score [199], GDT-TS [200], IS-Score [201], lDDT [202], DockQ [203], and pDockQ [157].”
De Novo Design of Cyclic Peptide Binders Based on Fragment Docking and Assembling
Zhang, C., Wang, F., Zhang, T., Yang, Y., Wang, L., Zhang, X., & Lai, L. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c00088 · Cited by 6
Estimating protein complex model accuracy using graph transformers and pairwise similarity graphs
Liu, J., Neupane, P., & Cheng, J. (2024) · Bioinformatics Advances · doi:10.1093/bioadv/vbaf180 · Cited by 6
“Each node represents an individual decoy, and an edge connects two decoys if they are similar according to a structural similarity metric such as TM-score, DockQ score (Basu and Wallner 2016), QS-score, and CAD-score (Olechnovi � c et al. 2013).”
Integrative modeling meets deep learning: Recent advances in modeling protein assemblies
Shor, B., & Schneidman‐Duhovny, D. (2024) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2024.102841 · Cited by 7
pyconsFold: a fast and easy tool for modeling and docking using distance predictions
Lamb, J., & Elofsson, A. (2021) · Bioinformatics · doi:10.1093/bioinformatics/btab353 · Cited by 6
“Replacing the interchain distances with predicted contacts significantly lowers the DockQ-score (Basu and Wallner, 2016), but the individual models TMscore does not decrease to the same extent, indicating that interchain contact predictions are less accurate.”
BackgroundInterPepScore: a deep learning score for improving the FlexPepDock refinement protocol
Johansson-Åkhe, I., & Wallner, B. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btac325 · Cited by 6
“InterPepScore is a graph neural network trained to predict the resulting DockQ score (Basu and Wallner, 2016) during FlexPepDock Monte Carlo sampling.”
Self-citationStudying protein–protein interaction through side-chain modeling method OPUS-Mut
Xu, G., Wang, Y., Wang, Q., & Ma, J. (2022) · Briefings in Bioinformatics · doi:10.1093/bib/bbac330 · Cited by 6
Data‐assisted protein structure modeling by global optimization in CASP12
Joo, K., Heo, S., Joung, I., Hong, S. H., Lee, S. J., & Lee, J. (2018) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.25457 · Cited by 6
“However, we found that 1 of the 5 models we submitted (model 3) was of relatively good docking pose between the 2 proteins with an interface RMSD of 6.99 Å calculated by DockQ.”
MethodInfluentialReliable protein-protein docking with AlphaFold, Rosetta, and replica-exchangePreprint
Harmalkar, A., Lyskov, S., & Gray, J. J. (2025) · eLife · doi:10.7554/elife.94029.2 · Cited by 11
Chitsazi, R., Wu, Y., participants, G. D. 2., Stevens, R. C., Zhao, S., & Kufareva, I. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.04.18.647407 · Cited by 10
“To complement our assessment of receptor-peptide complexes using the ‘correctness’ metric, we also sought to evaluate them by the DockQ score [96]: a metric for evaluating the accuracy of protein-protein/protein-peptide predictions developed in the context of the CAPRI competition [98].”
Physical-aware model accuracy estimation for protein complex using deep learning method
Wang, H., Sun, M., Xie, L., Liu, D., & Zhang, G. (2025) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2025.01.017 · Cited by 5
“In existing literature, most protein complex EMA methods use a distance threshold, such as 5 Å or 8 Å, between specific atoms (e.g., C β and C α ) to define the concept of contact[22,23].”
Topological deep learning for enhancing peptide-protein complex prediction
Dai, X., Wang, R., & Zhang, Y. (2025) · Communications Chemistry · doi:10.1038/s42004-025-01727-4 · Cited by 11
Jointly Embedding Protein Structures and Sequences through Residue Level Alignment
Birnbaum, F., Jain, S., Mądry, A., & Keating, A. E. (2024) · PRX Life · doi:10.1103/prxlife.2.043013 · Cited by 7
“We evaluate how well different methods discriminate between acceptable and decoy complexes (according to DockQ scores [39]).”
Shuvo, M. H., & Bhattacharya, D. (2025) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2024.12.015 · Cited by 6
Benchmarking TCR–pMHC structure prediction: a unified evaluation and CDR3-based functional insights
Lu, J. D., Zhu, X., Hu, X., Zhang, C., & Feng, F. (2026) · Briefings in Bioinformatics · doi:10.1093/bib/bbag289 · Cited by 6
Can we extract physics-like energies from generative protein diffusion Models?
Sarma, S., Truscott, H. H., Xu, D. Z., Reid, K. A., Chu, L., Chen, J. H. K., & Gray, J. J. (2026) · PRX Life · doi:10.1103/m7qr-gnjc · Cited by 5
Depetris, R. S., Lü, D., Polonskaya, Z., Zhang, Z., Luna, X., Tankard, A., Kolahi, P., Drummond, M. L., Williams, C., Ebert, M. C. C. J. C., et al. (2021) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.26280 · Cited by 5
“tools that evaluate the quality of docking models based on a single measure are a good set up for the application of machine learning methods on this assessment.(32,33) The mammalian vascular endothelial growth factors (VEGF-A,”
MethodChoi, S., Son, S. H., Kim, M. Y., Na, I., Uversky, V. N., & Kim, C. G. (2023) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2023.126526 · Cited by 6
Opuni, K. F. M., Ruß, M., Geens, R., Vocht, L. D., Wielendaele, P. V., Debuy, C., Sterckx, Y. G. ‐., & Glocker, M. O. (2024) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2024.08.023 · Cited by 5
Liu, S., Hao, X., Miao, D., & Zhang, Y. (2024) · The Journal of Physical Chemistry B · doi:10.1021/acs.jpcb.4c02325 · Cited by 6
Updates to the CASP Infrastructure in 2024
Kryshtafovych, A., Miłostan, M., Lensink, M. F., Velankar, S. S., Bonvin, A. M. J. J., Moult, J., & Fidelis, K. A. (2025) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.70042 · Cited by 6
Zarbafian, S., Moghadasi, M., Roshandelpoor, A., Nan, F., Li, K., Vakli, P., Vajda, S., Kozakov, D., & Paschalidis, I. C. (2018) · Scientific Reports · doi:10.1038/s41598-018-23982-3 · Cited by 5
Prates, N. S., Kamenik, A. S., Santos, L. H., Hofer, F., Liedl, K. R., & Ferreira, R. S. (2020) · Journal of Biomolecular Structure and Dynamics · doi:10.1080/07391102.2020.1796797 · Cited by 5
“The DockQ scores are obtained according to the equation (Basu & Wallner, 2016): DockQ Fnat , LRMS, iRMS, d1, d2ð Þ ¼ Fnat þ 1 1þ LRMSd1ð Þ2 þ 1 1þ iRMSd2ð Þ2 3 (1) To calculate the measures Fnat, LRMS, and iRMS, as defined by CAPRI (Lensink et al., 2007), the interface residues between receptor and…”
MethodHadi‐Alijanvand, H. (2019) · Journal of Proteome Research · doi:10.1021/acs.jproteome.9b00594 · Cited by 5
Predicted binding interface between coronavirus nsp3 and nsp4Preprint
Hensel, Z. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.03.05.483145 · Cited by 5
Co-translational assembly promotes functional diversification of paralogous proteinsPreprint
Mallik, S., Cisneros, A. F., Landry, C. R., & Levy, E. D. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.01.22.634331 · Cited by 6
Assessing scoring metrics for AlphaFold2 and AlphaFold3 protein complex predictions
Genz, L. R., Nair, S., Nagar, N., & Topf, M. (2025) · doi:10.1101/2025.04.16.648930 · Cited by 5
Protein language model embedded geometric graphs power inter-protein contact prediction
Si, Y., & Yan, C. (2023) · doi:10.1101/2023.01.07.523121 · Cited by 5
(2025) · doi:10.1016/j.cels.2025.101400 · Cited by 4
The rise of AlphaFold in drug design
Stecuła, A., Paul, R., Litchfield, K., Dalton, S. E., Low, C. M. R., Reis, C. R., & Congreve, M. (2025) · Progress in medicinal chemistry · doi:10.1016/bs.pmch.2025.09.002 · Cited by 7
Grandguillaume, I., da Silva, F. L. B., & Etchebest, C. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c01599 · Cited by 6
An interaction-derived graph learning framework for scoring protein–peptide complexes
Tao, H., Wang, X., & Huang, S. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-025-01136-1 · Cited by 7
“…such as quality assessments of decoys like the fraction of native contacts ( F nat ) 27 , the Local Distance Difference Test (LDDT) 48,49 and DockQ 50,51 , or classification of decoys as correct or incorrect based on interface root mean square deviation (iRMSD) or Critical Assessment of Predicted…”
Yuan, P., Chen, G., Li, Y., Liu, X., Saravanakumar, S., Zhao, J., Ji, Q., Wang, H., Lin, Y., Elbahnasawy, M. A., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-66322-6 · Cited by 5
Progress in structure prediction and design of adaptive immune receptors
Cohen, T., Hochner, T., & Schneidman‐Duhovny, D. (2026) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2026.103326 · Cited by 5
Integrative modeling of guanylate binding protein dimers
Schumann, W., Loschwitz, J., Reiners, J., Degrandi, D., Legewie, L., Stühler, K., Pfeffer, K., Poschmann, G., Smits, S. H. J., & Strodel, B. (2023) · Protein Science · doi:10.1002/pro.4818 · Cited by 4
“They are between 0.3 and 0.4, while only values between 0.8 and 1 (the possible maximum) indicate a similar protein – protein interface (Basu & Wallner, 2016).”
BackgroundInfluentialA significance score for protein–protein interaction models through random docking
McCoy, K. M., Ackerman, M. E., & Grigoryan, G. (2023) · Protein Science · doi:10.1002/pro.4853 · Cited by 4
“The current state of the art for evaluating PPI models is considered to be DOCKQ score (Basu & Wallner, 2016), which incorporates overall orientation but emphasizes the interface.”
MethodInfluentialThe Molecular Docking of MAX Fungal Effectors with Plant HMA Domain-Binding Proteins
Rozano, L., Hane, J. K., & Mancera, R. L. (2023) · International Journal of Molecular Sciences · doi:10.3390/ijms242015239 · Cited by 4
“A comparison of the docking poses to the native bound complexes was performed using DockQ [39], which provides a DockQ score between 0 and 1 (with 1 being closest to the native structure) based on the fraction of native contacts (fnat), interface RMSD (iRMSD), and ligand RMSD (lRMSD) [39].”
BackgroundMethodToward Characterising the Cellular 3D-Proteome
Elofsson, A. (2021) · Frontiers in Bioinformatics · doi:10.3389/fbinf.2021.598878 · Cited by 6
Improved protein docking by predicted interface residuesPreprint
Pozzati, G., Kundrotas, P. J., & Elofsson, A. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.22541/au.163254390.01798129/v1 · Cited by 4
Refinement of AlphaFold-Multimer structures with single sequence inputPreprint
Oda, T. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.12.27.521991 · Cited by 4
Metadata
- Authors
- Basu, S., & Wallner, B.
- Affiliations
- Linköping University (SE)
- Corresponding author
- Wallner, Björn
- Venue
- PLOS ONE
- Publisher
- Public Library of Science (PLoS)
- Published
- 2016-08-25
- Volume
- 11
- Issue
- 8
- Pages
- e0161879
- ISSN
- 1932-6203 (ISSN-L)
- PMC
- PMC4999177
- OpenAlex
- W2508425869
- Cited by
- 460
- References
- 26
- Citations by year
- 2026: 78 · 2025: 118 · 2024: 90 · 2023: 70 · 2022: 44 · 2021: 34 · 2020: 8 · 2019: 9 · 2018: 2 · 2017: 4 · 2016: 2
- FWCI
- 2.93
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- PLoS ONE (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DiVA (Linkoping University) (submittedVersion, other-oa)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Vetenskapsrådet (SE) (2012-5270)
- Swedish e-Science Research Center
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Structural Biology › Enzyme Structure and Function
- Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
- Keywords
- protein-protein docking, model ranking
- MeSH terms
- Algorithms, Protein Binding, Proteins, Reproducibility of Results, Computational Biology, Internet, Protein Interaction Mapping, Molecular Docking Simulation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.