Benchmarking AlphaFold2 on peptide structure prediction
Open accessAbstract
Recent advancements in computational tools have allowed protein structure prediction with high accuracy. Computational prediction methods have been used for modeling many soluble and membrane proteins, but the performance of these methods in modeling peptide structures has not yet been systematically investigated. We benchmarked the accuracy of AlphaFold2 in predicting 588 peptide structures between 10 and 40 amino acids using experimentally determined NMR structures as reference. Our results showed AlphaFold2 predicts α-helical, β-hairpin, and disulfide-rich peptides with high accuracy. AlphaFold2 performed at least as well if not better than alternative methods developed specifically for peptide structure prediction. AlphaFold2 showed several shortcomings in predicting Φ/Ψ angles, disulfide bond patterns, and the lowest RMSD structures failed to correlate with lowest pLDDT ranked structures. In summary, computation can be a powerful tool to predict peptide structures, but additional steps may be necessary to analyze and validate the results.
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
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?
McDonald et al benchmarked 588 peptides and found AF2 predicts many α-helical and β-hairpin peptides with surprising accuracy.
“McDonald et al performed a benchmark of 588 peptides revealing that AF2 predicts many α-helical and β-hairpin peptide structures with surprising accuracy25”
From The power and pitfalls of AlphaFold2 for structure prediction beyond rigid globular proteins (2024) · cites this paper as [25]
AF2 struggled with mixed secondary-structure membrane and soluble peptides, such as PRNP.
“AF2 was challenged by mixed secondary structure membrane and soluble peptides, such as the prion protein PRNP (Fig. 2a)25”
From The power and pitfalls of AlphaFold2 for structure prediction beyond rigid globular proteins (2024) · cites this paper as [25]
The top-ranked AF2 peptide models chosen by high pLDDT often did not have the lowest Cα RMSD to experimental structures, indicating pLDDT is not ideal for classifying peptide conformations.
“the best ranked AF2 models (selected on the basis of high pLDDT score) often did not exhibit the lowest Cα root-mean-square deviation (RMSD) relative to the experimental structure, suggesting the pLDDT metric used by AF2 to assess protein models is not optimal for classification of peptide conformations25”
From The power and pitfalls of AlphaFold2 for structure prediction beyond rigid globular proteins (2024) · cites this paper as [25]
Peptide structure prediction is difficult because AF2's training set omitted peptides, short sequences yield poor MSAs, and many peptides form conformational ensembles rather than one static structure.
“Peptide structure prediction poses additional challenges given that the benchmark set used to train AF2 excluded peptides, the difficulty in generating robust MSAs for short sequences, and observations that many peptides exist in solution as conformational ensembles rather than a single static conformation1,24,25”
From The power and pitfalls of AlphaFold2 for structure prediction beyond rigid globular proteins (2024) · cites this paper as [25]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
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 · see citing works
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 · see citing works
Harnessing protein folding neural networks for peptide–protein docking
Tsaban, T., Varga, J. K., Avraham, O., Ben-Aharon, Z., Khramushin, A., & Schueler-Furman, O. (2022) · Nature Communications · doi:10.1038/s41467-021-27838-9 · Cited by 1,205 · see citing works
De novo designed peptides for cellular delivery and subcellular localisation
Rhys, G. G., Cross, J. A., Dawson, W. M., Thompson, H. F., Shanmugaratnam, S., Savery, N. J., Dodding, M. P., Höcker, B., & Woolfson, D. N. (2022) · Nature Chemical Biology · doi:10.1038/s41589-022-01076-6 · Cited by 44 · see citing works
Template-free prediction of a new monotopic membrane protein fold and assembly by AlphaFold2
Gulsevin, A., Han, B., Porta, J. C., Mchaourab, H. S., Meiler, J., & Kenworthy, A. K. (2023) · Biophysical Journal · doi:10.1016/j.bpj.2022.11.011 · Cited by 16 · see citing works
Self-citation
Flagged first, then by citations
Research works printed but not linked: 28
- [1]Otvos L, and Wade JD (2014). Current challenges in peptide-based drug discovery. Front. Chem 2. PMC4126357 25152873
- [2]Toprak U (2020). The Role of Peptide Hormones in Insect Lipid Metabolism. Front. Physiol 11. PMC7221030 32457651
- [3]Hirakawa Y, and Sawa S (2019). Diverse function of plant peptide hormones in local signaling and development. Curr. Opin. Plant Biol 51, 81– 87. 10.1016/j.pbi.2019.04.005. 31132657 · DOI
- [4]Yeaman MR, and Yount NY (2003). Mechanisms of Antimicrobial Peptide Action and Resistance. Pharmacol. Rev 55, 27 LP – 55. 10.1124/pr.55.1.2. 12615953 · DOI
- [5]Lee ACL, Harris JL, Khanna KK, and Hong JH (2019). A comprehensive review on current advances in peptide drug development and design. Int. J. Mol. Sci 20, 1– 21. 10.3390/ijms20102383. PMC6566176 31091705 · DOI
- [6]Clark GC, Casewell NR, Elliott CT, Harvey AL, Jamieson AG, Strong PN, and Turner AD (2019). Friends or Foes? Emerging Impacts of Biological Toxins. Trends Biochem. Sci 44, 365– 379. 10.1016/j.tibs.2018.12.004. 30651181 · DOI
- [7]Luca S, Heise H, and Baldus M (2003). High-Resolution Solid-State NMR Applied to Polypeptides and Membrane Proteins. Acc. Chem. Res 36, 858– 865. 10.1021/ar020232y. 14622033 · DOI
- [8]Wüthrich K (2001). The way to NMR structures of proteins. Nat. Struct. Biol 8, 923– 925. 10.1038/nsb1101-923. 11685234 · DOI
- [9]Kuczera K (2015). Molecular Modeling of Peptides BT - Computational Peptidology In, Zhou P and Huang J, eds. (Springer New York), pp. 15– 41. 10.1007/978-1-4939-2285-7_2. · DOI
- [10]Thévenet P, Rey J, Moroy G, and Tuffery P (2014). De Novo Peptide Structure Prediction: An Overview. Comput. Pept 1268, 1– 13. 10.1007/978-1-4939-2285-7_1. 25555718 · DOI
- [11]Yan J, Bhadra P, Li A, Sethiya P, Qin L, Tai HK, Wong KH, and Siu SWI (2020). Deep-AmPEP30: Improve Short Antimicrobial Peptides Prediction with Deep Learning. Mol. Ther. - Nucleic Acids 20, 882– 894. 10.1016/j.omtn.2020.05.006. 32464552 PMC7256447 · DOI
- [12]Lamiable A, Thévenet P, Rey J, Vavrusa M, Derreumaux P, and Tufféry P (2016). PEP-FOLD3: faster de novo structure prediction for linear peptides in solution and in complex. Nucleic Acids Res 44, W449– 54. 10.1093/nar/gkw329. 27131374 PMC4987898 · DOI
- [13]Timmons PB, and Hewage CM (2021). APPTEST is a novel protocol for the automatic prediction of peptide tertiary structures. Brief. Bioinform 22, bbab308. 10.1093/bib/bbab308. 34396417 PMC8575040 · DOI
- [14]Minkyung B, Frank D, Ivan A, Justas D, Sergey O, Rie LG, Jue W, Qian C, N., K.L., Dustin SR, (2021). Accurate prediction of protein structures and interactions using a three-track neural network. Science (80-.) 373, 871– 876. 10.1126/science.abj8754. PMC7612213 34282049 · DOI · Paper page
- [15]Wu R, Ding F, Wang R, Shen R, Zhang X, Luo S, Su C, Wu Z, Xie Q, Berger B, (2022). High-resolution de novo structure prediction from primary sequence. bioRxiv, 2022.07.21.500999. 10.1101/2022.07.21.500999. · DOI · Paper page
- [16]Nerli S, McShan AC, and Sgourakis NG (2018). Chemical shift-based methods in NMR structure determination. Prog. Nucl. Magn. Reson. Spectrosc 106–107, 1– 25. 10.1016/j.pnmrs.2018.03.002. PMC6788782 31047599 · DOI
- [17]McDonald EF, Woods H, Smith ST, Kim M, Schoeder CT, Plate L, and Meiler J (2022). Structural Comparative Modeling of Multi-Domain F508del CFTR. Biomol 12. 10.3390/biom12030471. PMC8946492 35327663 · DOI
- [18]Gulsevin A, and Meiler J (2020). An investigation of three-finger toxin-nachr interactions through rosetta protein docking. Toxins (Basel) 12. 10.3390/toxins12090598. PMC7551183 32947868 · DOI
- [19]Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, Tunyasuvunakool K, Bates R, Žídek A, Potapenko A, (2021). Applying and improving AlphaFold at CASP14. Proteins Struct. Funct. Bioinforma 89, 1711– 1721. 10.1002/prot.26257. PMC9299164 34599769 · DOI
- [21]Johansson-Akhe I, and Wallner B (2021). Benchmarking Peptide-Protein Docking and Interaction Prediction with AlphaFold-Multimer. bioRxiv, 2021.11.16.468810.
- [22]Lee J (2021). C an a Lpha F Old 2 Predict Protein - Peptide Complex Structures Accurately ? bioRxiv, 2– 7.
- [23]Berman HM, Battistuz T, Bhat TN, Bluhm WF, Bourne PE, Burkhardt K, Feng Z, Gilliland GL, Iype L, Jain S, (2002). The Protein Data Bank. Acta Crystallogr. Sect. D 58, 899– 907. 10.1107/S0907444902003451. 12037327 · DOI
- [25]Fowler NJ, and Williamson MP (2022). The accuracy of protein structures in solution determined by AlphaFold and NMR. bioRxiv, 2022.01.18.476751. 10.1101/2022.01.18.476751. PMC9592556 35537451 · DOI
- [28]Leaver-Fay A, Tyka M, Lewis SM, Lange OF, Thompson J, Jacak R, Kaufman K, Renfrew PD, Smith CA, Sheffler W, (2011). Rosetta3: An object-oriented software suite for the simulation and design of macromolecules. Methods Enzymol 487, 545– 574. 10.1016/B978-0-12-381270-4.00019-6. 21187238 PMC4083816 · DOI
- [29]Monzon V, Haft DH, and Bateman A (2022). Folding the unfoldable: using AlphaFold to explore spurious proteins. Bioinforma. Adv 2, 1– 6. 10.1093/bioadv/vbab043. PMC9710616 36699409 · DOI
- [31]Xiong H, Han L, Wang Y, and Chai P (2022). Evaluating the Reliability of AlphaFold 2 for Unknown Complex Structures with Deep Learning. bioRxiv, 2022.07.08.499384. 10.1101/2022.07.08.499384. · DOI
- [32]Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN, and Bourne PE (2000). The Protein Data Bank. Nucleic Acids Res 28, 235– 242. 10.1093/nar/28.1.235. 10592235 PMC102472 · DOI · Paper page
- [34]Chaudhury S, Lyskov S, and Gray JJ (2010). PyRosetta: A script-based interface for implementing molecular modeling algorithms using Rosetta. Bioinformatics 26, 689– 691. 10.1093/bioinformatics/btq007. 20061306 PMC2828115 · DOI
Cited by
Sorted by citations
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 · see citing works
“McDonald et al. 25 performed a benchmark of 588 peptides, revealing that AF2 predicts many α-helical and β-hairpin peptide structures with surprising accuracy.”
BackgroundInfluentialMadaj, R., Martinez‐Goikoetxea, M., Kaminski, K., Ludwiczak, J., & Dunin‐Horkawicz, S. (2025) · Protein Science · doi:10.1002/pro.5244 · Cited by 23 · see citing works
Tikhonov, S., Babich, O., Sukhikh, S., Tikhonova, N., & Chernukha, I. (2025) · Journal of Applied Biology & Biotechnology · doi:10.7324/jabb.2025.219617 · Cited by 2 · see citing works
“At the same time, many new peptides have been developed, but only a few of them have regenerative properties [8].”
- Authors
- McDonald, E. F., Jones, T., Plate, L., Meiler, J., & Gulsevin, A.
- Affiliations
- Vanderbilt University (US)
- Corresponding author
- Meiler, Jens
- Gulsevin, Alican
- Venue
- Structure
- Publisher
- Elsevier BV
- Published
- 2023-01
- Volume
- 31
- Issue
- 1
- Pages
- 111-119.e2
- ISSN
- 0969-2126 (ISSN-L)
- PMC
- PMC9883802
- OpenAlex
- W4311539894
- Cited by
- 152
- References
- 5
- Citations by year
- 2026: 40 · 2025: 55 · 2024: 44 · 2023: 12 · 2022: 1
- FWCI
- 9.10
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- www.elsevier.com/tdm/userlicense/1.0/ (tdm)
- www.elsevier.com/legal/tdmrep-license (tdm)
- www.elsevier.com/open-access/userlicense/1.0/ (vor, 369 d)
- doi.org/10.15223/policy-017 (stm-asf)
- doi.org/10.15223/policy-037 (stm-asf)
- doi.org/10.15223/policy-012 (stm-asf)
- doi.org/10.15223/policy-029 (stm-asf)
- doi.org/10.15223/policy-004 (stm-asf)
- Locations
- Structure (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- Funders
- National Institutes of Health (R01 HL144131)
- National Institute of General Medical Sciences (1F31HL162483-01A1, R01 GM080403)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
- Physical Sciences › Materials Science › Materials Chemistry › Enzyme Structure and Function
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
- Keywords
- AlphaFold2, peptide structure prediction, benchmarking, NMR structure, alpha-helical peptides, β-hairpin peptides, disulfide-rich peptides, disulfide bond prediction, pLDDT, protein structure prediction
- MeSH terms
- Disulfides, Membrane Proteins, Peptides, Protein Conformation, Protein Structure, Secondary
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
