The DOI matches its Crossref record; 56 of the 56 references it cites were found in a registry or an index.
Impact of protein conformational diversity on AlphaFold predictions
Open accessPreprint — published as 10.1093/bioinformatics/btac202 · Paper page
Abstract
Abstract After the outstanding breakthrough of AlphaFold in predicting protein 3D models, new questions appeared and remain unanswered. The ensemble nature of proteins, for example, challenges the structural prediction methods because the models should represent a set of conformers instead of single structures. The evolutionary and structural features captured by effective deep learning techniques may unveil the information to generate several diverse conformations from a single sequence. Here we address the performance of AlphaFold2 predictions under this ensemble paradigm. Using a curated collection of apo-holo conformations, we found that AlphaFold2 predicts the holo form of a protein in 70% of the cases, being unable to reproduce the observed conformational diversity with an equivalent error than in the estimation of a single conformation. More importantly, we found that AlphaFold2’s performance worsens with the increasing conformational diversity of the studied protein. This impairment is related to the heterogeneity in the degree of conformational diversity found between different members of the homologous family of the protein under study. Finally, we found that main-chain flexibility associated with apo-holo pairs of conformers negatively correlates with the predicted local model quality score plDDT, indicating that plDDT values in a single 3D model could be used to infer local conformational changes linked to ligand binding transitions.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
In the citing paper
“Several preprints have provided evidence that AF2, despite its accuracy, likely does not learn the energy landscapes underpinning protein folding and function (Saldaño et al., 2021; Pak et al., 2021; Akdel et al., 2021).”
From Sampling alternative conformational states of transporters and receptors with AlphaFold2 (2022) · cites this paper as Saldaño et al., 2021
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
Highly accurate protein structure prediction with AlphaFold
Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., Tunyasuvunakool, K., Bates, R., Žídek, A., Potapenko, A., et al. (2021) · Nature · doi:10.1038/s41586-021-03819-2 · Cited by 43,764
ColabFold - Making protein folding accessible to all
Mirdita, M., Schütze, K., Moriwaki, Y., Heo, L., Ovchinnikov, S., & Steinegger, M. (2021) · doi:10.1101/2021.08.15.456425 · Cited by 9,391
Highly accurate protein structure prediction for the human proteome
Tunyasuvunakool, K., Adler, J., Wu, Z., Green, T., Zielinski, M., Žídek, A., Bridgland, A., Cowie, A., Meyer, C., Laydon, A., et al. (2021) · Nature · doi:10.1038/s41586-021-03828-1 · Cited by 3,025 · cited 2×
“Far from disappointing, this observation was expected since a large set of redundant protein structures (conformers) would have been required during the neural network training processes [5] in order to predict conformational diversity.”
BackgroundOpenMM 7: Rapid development of high performance algorithms for molecular dynamics
Eastman, P., Swails, J., Chodera, J. D., McGibbon, R. T., Zhao, Y., Beauchamp, K. A., Wang, L.-P., Simmonett, A. C., Harrigan, M. P., Stern, C. D., et al. (2017) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1005659 · Cited by 2,748
The relation between the divergence of sequence and structure in proteins.
Chothia, C., & Lesk, A. M. (1986) · The EMBO Journal · doi:10.1002/j.1460-2075.1986.tb04288.x · Cited by 2,458
The role of dynamic conformational ensembles in biomolecular recognition
Boehr, D. D., Nussinov, R., & Wright, P. E. (2009) · Nature Chemical Biology · doi:10.1038/nchembio.232 · Cited by 2,008
Anisotropy of Fluctuation Dynamics of Proteins with an Elastic Network Model
Atılgan, A. R., Durell, S. R., Jernigan, R. L., Demirel, M. C., Keskin, Ö., & Bahar, İ. (2001) · Biophysical Journal · doi:10.1016/s0006-3495(01)76033-x · Cited by 1,638
Large Amplitude Elastic Motions in Proteins from a Single-Parameter, Atomic Analysis
Tirion, M. M. (1996) · Physical Review Letters · doi:10.1103/physrevlett.77.1905 · Cited by 1,617
The ensemble nature of allostery
Motlagh, H. N., Wrabl, J. O., Li, J., & Hilser, V. J. (2014) · Nature · doi:10.1038/nature13001 · Cited by 1,247
THE KINETICS OF FORMATION OF NATIVE RIBONUCLEASE DURING OXIDATION OF THE REDUCED POLYPEPTIDE CHAIN
Anfinsen, C. B., Haber, E., Sela, M., & White, F. H. (1961) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.47.9.1309 · Cited by 1,182
Cited by
Sorted by citations
Predicting multiple conformations via sequence clustering and AlphaFold2
Wayment-Steele, H. K., Ojoawo, A., Otten, R., Apitz, J. M., Pitsawong, W., Hömberger, M., Ovchinnikov, S., Colwell, L., & Kern, D. (2024) · Nature · doi:10.1038/s41586-023-06832-9 · Cited by 545
“…has garnered interest in its ability to predict multiple conformations of proteins, yet AF2 has been demonstrated to fail in predicting multiple structures of metamorphic proteins 11 , proteins with apo/holo conformational changes 12 and other multi-state proteins 13 using its default settings.”
Sampling alternative conformational states of transporters and receptors with AlphaFold2
del Alamo, D., Sala, D., Mchaourab, H. S., & Meiler, J. (2022) · eLife · doi:10.7554/elife.75751 · Cited by 511
AlphaFold Meets Flow Matching for Generating Protein EnsemblesPreprint
Jing, B., Berger, B., & Jaakkola, T. (2024) · International Conference on Machine Learning · doi:10.48550/arxiv.2402.04845 · Cited by 293
SPEACH_AF: Sampling protein ensembles and conformational heterogeneity with Alphafold2
Stein, R. A., & Mchaourab, H. S. (2022) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1010483 · Cited by 258
Before and after AlphaFold2: An overview of protein structure prediction
Bertoline, L. M. F., Lima, A. N., Krieger, J. E., & Teixeira, S. K. (2023) · Frontiers in Bioinformatics · doi:10.3389/fbinf.2023.1120370 · Cited by 330
“In 67% of a dataset tested, AlphaFold prediction resembled holo form and the proteins were less predictable when the conformational differences between apo and holo forms increased (Saldaño et al., 2022).”
BackgroundModeling conformational states of proteins with AlphaFold
Sala, D., Engelberger, F., Mchaourab, H. S., & Meiler, J. (2023) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2023.102645 · Cited by 182
Easy and accurate protein structure prediction using ColabFold
Kim, G., Lee, S., Karin, E. L., Kim, H., Moriwaki, Y., Ovchinnikov, S., Steinegger, M., & Mirdita, M. (2024) · Nature Protocols · doi:10.1038/s41596-024-01060-5 · Cited by 170
AlphaFold predictions of fold-switched conformations are driven by structure memorization
Chakravarty, D., Schafer, J. W., Chen, E. A., Thole, J. F., Ronish, L. A., Lee, M., & Porter, L. L. (2024) · Nature Communications · doi:10.1038/s41467-024-51801-z · Cited by 159
“De fi ning “ Complex ” and “ Easy ” fold-switchers Since, the 92 fold-switching protein pairs are highly heterogeneous with respect to the structural differences between conformers, a fair comparisonofthepredictionsproducedbyAF2samplingswouldneed a separation of the targets based on their dif fi culty or complexity.”
InfluentialAlphaFold, Artificial Intelligence (AI), and Allostery
Nussinov, R., Zhang, M., Liu, Y., & Jang, H. (2022) · The Journal of Physical Chemistry B · doi:10.1021/acs.jpcb.2c04346 · Cited by 190
The power and pitfalls of AlphaFold2 for structure prediction beyond rigid globular proteins
Agarwal, V., & McShan, A. C. (2024) · Nature Chemical Biology · doi:10.1038/s41589-024-01638-w · Cited by 155
“The standard implementation of AlphaFold performs well only in detecting a single structural snapshot (the ‘ground state’ structure), likely due to the lack of a large set of redundant protein conformers in the training set 16,32 .”
BackgroundEigenFold: Generative Protein Structure Prediction with Diffusion ModelsPreprint
Jing, B., Erives, E., Pao-Huang, P., Corso, G., Berger, B., & Jaakkola, T. (2023) · PubMed · doi:10.48550/arxiv.2304.02198 · Cited by 36
Accelerating Cryptic Pocket Discovery Using AlphaFold
Meller, A., Bhakat, S., Solieva, S. O., & Bowman, G. R. (2023) · Journal of Chemical Theory and Computation · doi:10.1021/acs.jctc.2c01189 · Cited by 136
“In other cases, like with plasmepsin II, AlphaFold samples partially open states (Figure 2C).”
AFsample2 predicts multiple conformations and ensembles with AlphaFold2
Kalakoti, Y., & Wallner, B. (2025) · Communications Biology · doi:10.1038/s42003-025-07791-9 · Cited by 126
“Inaddition, it has previously been reported that the model con fi dence of AF2 predictions deteriorates with increased subsampling 18 .”
Evaluation of AlphaFold2 structures as docking targets
Holcomb, M., Chang, Y., Goodsell, D. S., & Forli, S. (2022) · Protein Science · doi:10.1002/pro.4530 · Cited by 112
“…power of AF2-predicted structures in the context of a smaller set of 38 proteins (Zhang et al., 2022), and with a study that suggests AF2 has a tendency to predict holo structures (Saldaño et al., 2022), and in the case of GPCRs can be biased to produce the desired conformation (Heo & Feig, 2022).”
BackgroundProtein Conformation Generation via Force-Guided SE(3) Diffusion ModelsPreprint
Wang, Y., Wang, L., Shen, Y., Wang, Y., Yuan, H., Wu, Y., & Gu, Q. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2403.14088 · Cited by 43
Str2Str: A Score-based Framework for Zero-shot Protein Conformation SamplingPreprint
Lu, J., Zhong, B., Zhang, Z., & Tang, J. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2306.03117 · Cited by 35
“…models (Baek et al., 2021; Jumper et al., 2021; Lin et al., 2023) which enables the study of proteins based on high-accuracy structures, the predicted ensembles often lack diversity (Chakravarty & Porter, 2022; Salda˜no et al., 2022) and modeling structure-dynamics relationship remains a challenge.”
BackgroundNovel machine learning approaches revolutionize protein knowledge
Bordin, N., Dallago, C., Heinzinger, M., Kim, S., Littmann, M., Rauer, C., Steinegger, M., Rost, B., & Orengo, C. (2023) · Trends in Biochemical Sciences · doi:10.1016/j.tibs.2022.11.001 · Cited by 72
“The sudden availability of a reliable model relieved many research groups from long-term structural characterization efforts, allowing for targeted answers in conformational studies [94,95], oligomerization prediction [96,97], drug channel conformations [98], and early-stage assembly of complexes in disease [99].”
BackgroundPrediction of multiple conformational states by combining sequence clustering with AlphaFold2Preprint
Wayment-Steele, H. K., Ovchinnikov, S., Colwell, L. J., & Kern, D. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.10.17.512570 · Cited by 66
Deep learning for protein structure prediction and design—progress and applications
Jänes, J., & Beltrão, P. (2024) · Molecular Systems Biology · doi:10.1038/s44320-024-00016-x · Cited by 76
“Early studies suggested that AlphaFold2 could not predict the structural changes of mutated sequences (Buel and Walters, 2022) or different conformation of proteins are known to change in structure when bound to a small molecule (Saldaño et al, 2022).”
BackgroundLee, M., Schafer, J. W., Prabakaran, J., Chakravarty, D., Clore, M. F., & Porter, L. L. (2025) · Nature Communications · doi:10.1038/s41467-025-60759-5 · Cited by 38
“-no novel biological insight such as blind predictions of unknown folds at scale and how reliable the predictions are likely/ what is the confidence?”
P2DFlow: A Protein Ensemble Generative Model with SE(3) Flow Matching
Jin, Y., Huang, Q., Song, Z.-Y., et al. (2024) · Journal of Chemical Theory and Computation · doi:10.48550/arxiv.2411.17196 · Cited by 38
PINDER: The protein interaction dataset and evaluation resourcePreprint
Kovtun, D., Akdel, M., Goncearenco, A., Zhou, G., Holt, G. T., Baugher, D., Lin, D., Adeshina, Y. O., Castiglione, T., Wang, X., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.07.17.603980 · Cited by 37
“10, that predicted structures seem to have a better performance than apo structures, likely due to the fact that AlphaFold tends to predict structures closer to the holo conformation (Salda˜no et al., 2022).”
BackgroundResultStructure Language Models for Protein Conformation GenerationPreprint
Lu, J., Chen, X., Lu, S. Z., Shi, C., Guo, H., Bengio, Y., & Tang, J. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2410.18403 · Cited by 15
“…et al. (2023) have curated two benchmarking sets of pairing data including (1) 77 pairs of fold-switching proteins (Chakravarty & Porter, 2022); and (2) 90 apo/holo pairs with ligand-induced conformational change (Salda˜no et al., 2022) to evaluate the modeling capacity of conformation diversity.”
ProteinBench: A Holistic Evaluation of Protein Foundation ModelsPreprint
Ye, F., Zheng, Z., Xue, D., Shen, Y., Wang, L., Ma, Y., Wang, Y., Wang, X., Zhou, X., & Gu, Q. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2409.06744 · Cited by 31
“The choice of model should be carefully aligned with the intended applications; ( 3 ) While generative models extended from classic folding models have shown the ability to sample protein conformations, challenges remain in both multiple-state prediction and distribution prediction.”
Generative AI in depth: A survey of recent advances, model variants, and real-world applications
Yazdani, S., Singh, A., Saxena, N., Wang, Z., Palikhe, A., Pan, D., Pal, U., Yang, J., & Zhang, W. (2025) · Journal of Big Data · doi:10.1186/s40537-025-01247-x · Cited by 40
Malhotra, Y., John, J., Yadav, D., Sharma, D., Vanshika, Rawal, K., Mishra, V., & Chaturvedi, N. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.109842 · Cited by 42
Lazou, M., Khan, O., Nguyen, T., Padhorny, D., Kozakov, D., Joseph‐McCarthy, D., & Vajda, S. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2412719121 · Cited by 37
Agajanian, S., Alshahrani, M., Bai, F., Tao, P., & Verkhivker, G. M. (2023) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.2c01634 · Cited by 37
Representing structures of the multiple conformational states of proteins
Ramelot, T. A., Tejero, R., & Montelione, G. T. (2023) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2023.102703 · Cited by 26
“Subsequently, the proclivity of AF2 to model one of multiple conformational states was also reported for a collection of ~100 apo / holo protein structure pairs [51].”
BackgroundSimpleFold: Folding Proteins is Simpler than You ThinkPreprint
Wang, Y.-Y., Lu, J.-R., Jaitly, N., Susskind, J. M., & Bautista, M. A. (2025) · International Conference on Learning Representations · doi:10.48550/arxiv.2509.18480 · Cited by 33
“3, SimpleFold obtains state-of-the-art performance on Apo/holo , where SimpleFold outperforms strong MSA-based approaches like AlphaFlow significantly.”
AFsample2: Predicting multiple conformations and ensembles with AlphaFold2Preprint
Kalakoti, Y., & Wallner, B. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.05.28.596195 · Cited by 26
Ellaway, J. I. J., Anyango, S., Nair, S. S., Zaki, H. A., Nadzirin, N., Powell, H. R., Gutmanas, A., Váradi, M., & Velankar, S. S. (2024) · Structural Dynamics · doi:10.1063/4.0000251 · Cited by 28
Recent Advances and Challenges in Protein Structure Prediction
Peng, C., Liang, F., Xia, Y., Zhao, K., Hou, M., & Zhang, G. (2023) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.3c01324 · Cited by 42
Aubel, M., Eicholt, L. A., & Bornberg‐Bauer, E. (2023) · F1000Research · doi:10.12688/f1000research.130443.1 · Cited by 24
Deep generative modeling of temperature-dependent structural ensembles of proteins
Janson, G., Jussupow, A., & Feig, M. (2025) · Communications Chemistry · doi:10.1038/s42004-025-01737-2 · Cited by 33
“We further analyzedhomologousproteins with different experimental T m to explore whether a generative model can probe sequence properties in fl uencing thermal stability 47 , focusing on fi ve protein pairs 48 (Supplementary Table 6).”
InfluentialRandom, de novo, and conserved proteins: How structure and disorder predictors perform differently
Middendorf, L., & Eicholt, L. A. (2024) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.26652 · Cited by 24
“44,46,48 We here conclude that such kinds of analyses are limited for de novo and random proteins, and their results from structure predictions differ highly from those of conserved proteins.”
BackgroundNeuralPLexer3: Accurate Biomolecular Complex Structure Prediction with Flow ModelsPreprint
Qiao, Z., Ding, F., Dresselhaus, T., Rosenfeld, M. A., Han, X., Howell, O., Iyengar, A., Opalenski, S., Christensen, A. S., Sirumalla, S. K., et al. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2412.10743 · Cited by 16
Olanders, G., Testa, G., Tibo, A., Nittinger, E., & Tyrchan, C. (2024) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.4c01475 · Cited by 27
Meller, A., de Oliveira, S. H. P., Davtyan, A., Abramyan, T. M., Bowman, G. R., & van den Bedem, H. (2023) · Frontiers in Molecular Biosciences · doi:10.3389/fmolb.2023.1171143 · Cited by 28
“While our work was based off a single AF structure as a starting point, we are aware of efforts to use these DL protein structure prediction tools to sample multiple conformations, thus better capturing protein flexibility (Saldanõ et al., 2022; Meller et al., 2023a).”
Generation of protein dynamics by machine learning
Janson, G., & Feig, M. (2025) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2025.103115 · Cited by 21
Exploring protein functions from structural flexibility using CABS ‐flex modeling
Nithin, C., Fornari, R. P., Pilla, S. P., Wróblewski, K., Zalewski, M., Madaj, R., Koliński, A., Macnar, J. M., & Kmiecik, S. (2024) · Protein Science · doi:10.1002/pro.5090 · Cited by 21
“However, these endeavors are not without challenges, such as reconciling the static nature of structural predictions with the inherently dynamic nature of proteins (Saldaño et al., 2022) or better prediction of protein regions that are disordered under ambient conditions (Ma et al., 2023).”
BackgroundAlphaFold2 as a replacement for solution NMR structure determination of small proteins: Not so fast!
Bonin, J. P., Aramini, J. M., Dong, Y., Wu, H., & Kay, L. E. (2024) · Journal of Magnetic Resonance · doi:10.1016/j.jmr.2024.107725 · Cited by 21
“In cases where a protein adopts multiple conformers, AF2 typically predicts only one of the structures, failing to capture the conformational diversity seen in the protein [19].”
Automated benchmarking of combined protein structure and ligand conformation prediction
Leemann, M., Sagasta, A., Eberhardt, J., Schwede, T. F., Robin, X., & Durairaj, J. (2023) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.26605 · Cited by 24
“Despite AlphaFold being shown to prefer prediction of the holo form, 21 homology modeling was recently shown to be better at the PLC prediction task.”
Investigating the conformational landscape of AlphaFold2-predicted protein kinase structures
Masri, C. A., Trozzi, F., Lin, S.-H., Tran, O., Sahni, N., Pátek, M., Cichońska, A., Ravikumar, B., & Rahman, R. (2023) · Bioinformatics Advances · doi:10.1093/bioadv/vbad129 · Cited by 24
“Heged } us et al. (2022), Guo et al. (2022), and Salda ~ no et al. (2022) have also shown a relationship between pLDDT and protein conformational dynamics.”
Protein Structure Tokenization: Benchmarking and New Recipe
Yuan, X., Wang, Z., Collins, M., & Rangwala, H. (2025) · arXiv · doi:10.48550/arxiv.2503.00089
“3 shows that: (1) AminoAseed outperforms all models in Apo Holo and Fold Switching datasets, with an average relative gain of +12.83% over ESM3 across all metrics; (2) ESM3 performs second best; (3) VanillaVQ slightly trails ESM3 by -0.82%; (4) IF-based PSTs underperform most VQ-VAE-based PSTs because their training o…”
InfluentialAlphaFold2 can predict single-mutation effects
McBride, J. M., Polev, K., Abdirasulov, A., Reinharz, V., Grzybowski, B. A., & Tlusty, T. (2022) · doi:10.1101/2022.04.14.488301 · Cited by 87
Spoendlin, F. C., Fernández-Quintero, M. L., Raghavan, S. S. R., Turner, H. L., Gharpure, A., Loeffler, J. R., Wong, W. K., Bujotzek, A., Georges, G., Ward, A. B., et al. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-025-01131-6 · Cited by 20
“22,23,26–30).”
Deep learning methods for protein structure prediction
Qin, Y., Chen, Z., Ye, P., Xiao, Y., Zhong, T., & Yu, X. (2024) · MedComm – Future Medicine · doi:10.1002/mef2.96 · Cited by 33
Challenges in describing the conformation and dynamics of proteins with ambiguous behavior
Roca‐Martínez, J., Lázár, T., Gavaldá-García, J., Bickel, D., Pancsa, R., Dixit, B., Tzavella, K., Ramasamy, P., Sanchez-Fornaris, M., Grau, I., et al. (2022) · Frontiers in Molecular Biosciences · doi:10.3389/fmolb.2022.959956 · Cited by 20
“Indeed, in relation to conformational diversity as observed in the PDB from apo-holo pairs of conformers for the same protein (Saldaño et al., 2022), AlphaFold2 predicts the holo form in ~70% of cases but is unable to capture both states.”
BackgroundReliable protein-protein docking with AlphaFold, Rosetta, and replica-exchange
Harmalkar, A., Lyskov, S., & Gray, J. J. (2023) · doi:10.1101/2023.07.28.551063 · Cited by 40
Modeling Alternate Conformations with Alphafold2 via Modification of the Multiple Sequence Alignment
Stein, R. A., & Mchaourab, H. S. (2021) · doi:10.1101/2021.11.29.470469 · Cited by 258
“A recent report examining a curated subset of apo and ligand bound structures 302 suggested that pLDDT values infer local conformational changes.(31) The results presented here do not 303 suggest that these regions of low confidence are incorrect in their two- or three-dimensional structure, 304 but appear to be regio…”
BackgroundKersten, C., Clower, S., & Barthels, F. (2023) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.2c01400 · Cited by 15
Deep learning approaches for conformational flexibility and switching properties in protein design
Rudden, L. S. P., Hijazi, M., & Barth, P. (2022) · Frontiers in Molecular Biosciences · doi:10.3389/fmolb.2022.928534 · Cited by 12
“However, it has been indicated that AlphaFold2 is unable to accurately learn the underlying energy landscape that describes protein folding and function (Saldaño et al., 2021).”
BackgroundHidden Structural States of Proteins Revealed by Conformer Selection with AlphaFold-NMR
Huang, Y., Ramelot, T. A., Spaman, L. E., et al. (2024) · bioRxiv · doi:10.21203/rs.3.rs-5994356/v1 · Cited by 13
“However, for highly-flexible proteins like GLuc, these differences highlight AF2's limitations in capturing the solution-dominant states 11,29,35-38 .”
Raisinghani, N., Alshahrani, M., Gupta, G., Tian, H., Xiao, S., Tao, P., & Verkhivker, G. M. (2024) · Journal of Chemical Theory and Computation · doi:10.1021/acs.jctc.4c00222 · Cited by 17
Verkhivker, G. M., Alshahrani, M., Gupta, G., Xiao, S., & Tao, P. (2023) · International Journal of Molecular Sciences · doi:10.3390/ijms24097747 · Cited by 16
“Several recent studies have provided strong evidence that structure prediction capabilities of the AF2 methods are not trivially expandable for the prediction of complex allosteric landscapes and mechanisms of allosteric regulation [69,70].”
BackgroundUrban, P., & Pompon, D. (2022) · Scientific Reports · doi:10.1038/s41598-022-20390-6 · Cited by 15
Quantification of the Impact of Structure Quality on Predicted Binding Free Energy Accuracy
Behera, S., Hahn, D. F., Wilson, C. J., Marsili, S., Tresadern, G., Gapsys, V., & de Groot, B. L. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c00947 · Cited by 14
Shen, C., Han, X., Cai, H., Chen, T., Kang, Y., Pan, P., Ji, X., Hsieh, C., Deng, Y., & Hou, T. (2025) · Journal of Medicinal Chemistry · doi:10.1021/acs.jmedchem.4c02740 · Cited by 10
Rustamov, K., & Baev, A. (2024) · Bioinformatics Advances · doi:10.1093/bioadv/vbae197 · Cited by 13
“…these advancements, AF2’s initial limitation was its inability to predict different substates of metamorphic proteins (Chakravarty and Porter 2022, Salda ~ no et al. 2022), a challenge that has been recently addressed by various multiple sequence alignment (MSA) clustering approaches (Monteiro da…”
InfluentialRaisinghani, N., Parikh, V., Foley, B., & Verkhivker, G. M. (2024) · International Journal of Molecular Sciences · doi:10.3390/ijms252312968 · Cited by 10
Limitations of Protein Structure Prediction Algorithms in Therapeutic Protein Development
Niazi, S. K., Mariam, Z., & Paracha, R. Z. (2024) · BioMedInformatics · doi:10.3390/biomedinformatics4010007 · Cited by 20
Sampling the conformational landscapes of transporters and receptors with AlphaFold2
del Alamo, D., Sala, D., Mchaourab, H. S., & Meiler, J. (2021) · doi:10.1101/2021.11.22.469536 · Cited by 511
“In contrast with a recent preprint 35 , the predicted flexibility values failed to correlate with their pLDDT values, which reflect the confidence of the AF2 prediction of each residue’s local environment 36 .”
BackgroundBerksoz, M., & Atilgan, C. (2024) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.26688 · Cited by 8
“40 Models also showed varying degrees of similarity to the holo sensor structure (PDB: 3OSR).”
BackgroundRamasamy, P., Zuallaert, J., Martens, L., & Vranken, W. F. (2026) · Protein Science · doi:10.1002/pro.70376 · Cited by 15
Rotem‐Bamberger, S., Fahoum, J., Keinan‐Adamsky, K., Tsaban, T., Avraham, O., Shalev, D. E., Chill, J. H., & Schueler‐Furman, O. (2022) · Journal of Biological Chemistry · doi:10.1016/j.jbc.2022.102145 · Cited by 14
Toward the Prediction of Binding Events in Very Flexible, Allosteric, Multidomain Proteins
Basciu, A., Athar, M., Kurt, H., Neville, C., Malloci, G., Muredda, F. C., Bosin, A., Ruggerone, P., Bonvin, A. M. J. J., & Vargiu, A. V. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.4c01810 · Cited by 12
The Rapid Evolution of De Novo Proteins in Structure and Complex
(2024) · doi:10.1093/gbe/evae107 · Cited by 13
Piomponi, V., Cazzaniga, A., & Cuturello, F. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.04.08.647782 · Cited by 1
“Nevertheless, despite its remarkable performance, AF2 is primarily designed to predict a single dominant conformation, and its ability to capture multiple functional states of a protein remains limited [2,3].”
Hashem, S., Dougha, A., & Tufféry, P. (2025) · Molecules · doi:10.3390/molecules30051047 · Cited by 18
Learning to engineer protein flexibility
Kouba, P., Planas-Iglesias, J., Damborský, J., et al. (2024) · International Conference on Learning Representations · doi:10.48550/arxiv.2412.18275 · Cited by 10
“…be induced by sequence variability in the region, for instance), it has been shown to be negatively correlated with flexibility (Ruff & Pappu, 2021; Guo et al., 2022; Salda ˜ no et al., 2022) and exploited to approximate flexibility as measured by Nuclear Magnetic Resonance (NMR) (Ma et al., 2023).”
Multi-state Protein Design with DynamicMPNNPreprint
Abrudan, A., Ojeda, S. P., Joshi, C. K., Greenig, M., Engelberger, F., Khmelinskaia, A., Meiler, J., Vendruscolo, M., & Knowles, T. P. J. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2507.21938 · Cited by 11
“For dataset splitting, we first curate a benchmark set from five previous studies of proteins with large conformational changes: 92 metamorphic proteins (Porter & Looger, 2018), 91 apo-holo proteins (Salda˜no et al., 2022 Overall architecture .”
Swapna, G. V. T., Dube, N., Roth, M. J., & Montelione, G. T. (2025) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1013590 · Cited by 9
“Recent advances have been reported using modified AF2 protocols and “enhanced sampling” methods to model multiple conformational states of proteins, including integral membrane proteins [14].”
Renthal, R. D. (2024) · Current Research in Insect Science · doi:10.1016/j.cris.2024.100082 · Cited by 8
“The favoring by AF of the open (or apo) conformation contrasts with a recent survey of 91 apo/holo pairs of AF models that found 67 % of the structures were similar to the holo (ligand-bound) structure (Salda ˜ no et al., 2022).”
Feidakis, C. P., Krivák, R., Hoksza, D., & Novotný, M. (2024) · Journal of Molecular Biology · doi:10.1016/j.jmb.2024.168545 · Cited by 10
Zlobin, A. S., & Golovin, A. V. (2022) · ACS Omega · doi:10.1021/acsomega.2c05201 · Cited by 11
Ferrario, E., Miggiano, R., Rizzi, M., & Ferraris, D. M. (2022) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2022.07.027 · Cited by 11
“It has been previously observed that AF favours the prediction of the ligand-bound rather than the ligand-free protein conformations [35].”
BackgroundHow to select the best model from AlphaFold2 structures?Preprint
Takei, Y., & Ishida, T. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.04.05.487218 · Cited by 11
“Holo -structures are more likely to be output than apo-structures when multiple prediction models are used[19].”
BackgroundAccelerated sampling of protein dynamics using BioEmu augmented molecular simulationPreprint
Bhakat, S., & Strauch, E.-M. (2026) · bioRxiv · doi:10.64898/2026.01.07.698041 · Cited by 9
Biała‐Leonhard, W., Bigos, A., Brezovský, J., & Jasiński, M. (2025) · PLANT PHYSIOLOGY · doi:10.1093/plphys/kiaf146 · Cited by 8
Raisinghani, N., Alshahrani, M., Gupta, G., & Verkhivker, G. M. (2024) · International Journal of Molecular Sciences · doi:10.3390/ijms251810082 · Cited by 7
Computational design of Periplasmic binding protein biosensors guided by molecular dynamics
O’Shea, J. M., Richardson, A., Doerner, P. W., & Wood, C. W. (2024) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1012212 · Cited by 7
“…Coefficient matrix can be accessed from only the holo crystal structure, and could even be accessed for PBPs with no known crystal structure thanks to structural prediction algorithms such as AlphaFold2, which are known to mostly predict holo states of binding proteins (Saldaño et al. , 2022).”
BackgroundZhao, H.-F., Zhang, H., She, Z., Gao, Z., Wang, Q., Geng, Z. Z., & Dong, Y. (2023) · International Journal of Molecular Sciences · doi:10.3390/ijms24032740 · Cited by 12
“Moreover, AF2 generally provides a single static picture of a protein structure that it thinks is most likely to appear in the PDB, whereas proteins exist in many different conformations, and AF2 currently cannot capture all aspects of a protein’s biological function [61,62].”
BackgroundLi, H., Nithin, C., Kmiecik, S., & Huang, S. (2025) · Drug Discovery Today · doi:10.1016/j.drudis.2025.104382 · Cited by 9
Robust Prediction of Multiple Protein Conformations with Entropy GuidancePreprint
Wu, D. T., & Feng, L. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.04.26.650728 · Cited by 7
Li, J., Zhu, Z., & Song, C. (2026) · Quantitative Biology · doi:10.1101/2024.06.09.598121 · Cited by 6
“13 Predicting alternative conformations and assessing the impact of sequence mutations still pose difficulties with standard AF2 usage.”
BackgroundMethodInfluentialTRPA5 encodes a thermosensitive ankyrin ion channel receptor in a triatomine insect
Liénard, M. A., Báez-Nieto, D., Tsai, C., Valencia‐Montoya, W. A., Werin, B., Johanson, U., Lassance, J., Pan, J. Q., Yu, N., & Pierce, N. E. (2024) · iScience · doi:10.1016/j.isci.2024.109541 · Cited by 8
Rueda, A. J. V., Bulgarelli, F. L., Palópoli, N., & Parisi, G. D. (2023) · Database · doi:10.1093/database/baad010 · Cited by 7
Self-citationLombard, V., Van, J. N., Grudinin, S., & Laine, É. (2026) · Acta Crystallographica Section D Structural Biology · doi:10.1107/s2059798326006054 · Cited by 4
“Beyond the single-structure frontier, several studies have underscored the limitations and potential of protein structure predictors (PSP) for generating alternative conformations (Salda˜no et al., 2022; Lane, 2023; Bryant & No´e, 2024; Chakravarty et al., 2025).”
Prediction of structural variation
Kalakoti, Y., Sanjeev, A., & Wallner, B. (2025) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2025.103003 · Cited by 8
Raisinghani, N., Alshahrani, M., Gupta, G., Tian, H., Xiao, S., Tao, P., & Verkhivker, G. M. (2024) · The Journal of Physical Chemistry B · doi:10.1021/acs.jpcb.4c04985 · Cited by 4
Tang, Y., Yu, M., Bai, G., Li, X., Xu, Y., & Ma, B. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.06.27.600251 · Cited by 5
“All these methods are developed and trained as deterministic maps from input (sequence or MSA) to output (structure), making them suboptimal for modeling structural ensembles 9 .”
MethodModeling with Alternate Locations in X-ray Protein Structures
Gutermuth, T., Sieg, J., Stohn, T., & Rarey, M. (2023) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.3c00100 · Cited by 6
Sawhney, A., Li, J., & Liao, L. (2024) · International Journal of Molecular Sciences · doi:10.3390/ijms25105247 · Cited by 6
“Several studies have examined AlphaFold2’s predicted structures, for example to assess the impact of conformational diversity on its predictions [29] or to evaluate if AlphaFold2 learned the physics of folding [30].”
BackgroundExploring the potential of AlphaFold distograms for predicting binding‐induced hinge motions
Savaş, B., Barlas, A. B., & Karaca, E. (2026) · FEBS Letters · doi:10.1002/1873-3468.70297 · Cited by 6
Jurgelaitis, E., Zagorskaitė, E., Kopu̅stas, A., et al. (2025) · Cell Research · doi:10.1038/s41422-025-01198-1 · Cited by 5
“57 Since the beta-relay involves signi fi cant rearrangement of the local structural environment, AlphaFold might be expected to assign lower con fi dence values to the corresponding region.”
Kim, T. D., Pretorius, D., Murray, J. W., & Cardona, T. (2025) · Physiologia Plantarum · doi:10.1111/ppl.70284 · Cited by 4
“Several studies have found that the pLDDT score is a good indicator of the flexibility of the region, with lower values indicating more dynamicity (Guo et al. 2022; Saldano et al. 2022; Ma et al. 2023).”
JAMUN: Bridging Smoothed Molecular Dynamics and Score-Based Learning for Conformational EnsemblesPreprint
Daigavane, A., Vani, B. P., Davidson, D., Saremi, S., Rackers, J. A., & Kleinhenz, J. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2410.14621 · Cited by 3
“Protein structure prediction models tend to struggle to capture this diversity, especially when sampling the conformations of flexible domains [79, 17, 2] and rarer conformations [56] not found in the Protein Data Bank (PDB) [13].”
Drake, L. A., Hahn, A., Dixon, A., et al. (2023) · Journal of Biological Chemistry · doi:10.1016/j.jbc.2023.104869 · Cited by 5
Metadata
- Authors
- Saldaño, T., Escobedo, N., Marchetti, J., Zea, D. J., Mac Donagh, J., Velez Rueda, A. J., Gonik, E., Melani, A. G., Novomisky Nechcoff, J., Salas, M. N., Peters, T., Demitroff, N., Fernandez Alberti, S., Palopoli, N., Fornasari, M. S., & Parisi, G.
- Affiliations
- National University of Quilmes (AR)
- Consejo Nacional de Investigaciones Científicas y Técnicas (AR)
- Fundación Instituto Leloir (AR)
- Corresponding author
- Parisi, Gustavo
- ORCID
- Venue
- Bioinformatics
- Publisher
- openRxiv
- Published
- 2021-10-28
- Volume
- 38
- Issue
- 10
- Pages
- 2742-2748
- ISSN
- 1367-4803 (ISSN-L)
- OpenAlex
- W4225552973
- Cited by
- 196
- References
- 55
- Citations by year
- 2026: 45 · 2025: 54 · 2024: 52 · 2023: 29 · 2022: 16
- FWCI
- 12.32
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Licence
- cc-by
- Locations
- Bioinformatics (publishedVersion)
- bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by)
- PubMed (publishedVersion)
- Greater South Information System
- Funders
- European Commission (HORIZON2020, 778247)
- Consejo Nacional de Investigaciones Científicas y Técnicas (PIP-2015-2017 11220150100853CO, PICT2018)
- Universidad Nacional de Quilmes (PUNQ 1309/19, 778247)
- Agencia Nacional de Promoción de la Investigación, el Desarrollo Tecnológico y la Innovación (PICT-2018 3457)
- Agencia Nacional de Promoción Científica y Tecnológica (778247, CONICET)
- 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 › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
- Keywords
- protein conformational diversity, AlphaFold2, ColabFold, protein ensembles, ligand binding, pLDDT, model quality assessment, homologous protein families, protein structure prediction
- MeSH terms
- Protein Binding, Protein Conformation, Proteins
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
