CoChat
Sign Up
Cited by 5Protein Structure and DynamicsMachine Learning in BioinformaticsEnzyme Structure and Function

The DOI matches its Crossref record; 3 of the 3 references it cites were found in a registry or an index.

TAFPred: Torsion Angle Fluctuations Prediction from Protein Sequences

Open access
Kabir, Md Wasi Ul, Alawad, Duaa Mohammad, Mishra, Avdesh, Hoque, Md Tamjidul·2023·Biology
DOI 10.3390/biology12071020PMC PMC10376102

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Protein molecules show varying degrees of flexibility throughout their three-dimensional structures. The flexibility is determined by the fluctuations in torsion angles, specifically phi (φ) and psi (ψ), which define the protein backbone. These angle fluctuations are derived from variations in backbone torsion angles observed in different models. By analyzing the fluctuations in Cartesian coordinate space, we can understand the structural flexibility of proteins. Predicting torsion angle fluctuations is valuable for determining protein function and structure when these angles act as constraints. In this study, a machine learning method called TAFPred is developed to predict torsion angle fluctuations using protein sequences directly. The method incorporates various features, such as disorder probability, position-specific scoring matrix profiles, secondary structure probabilities, and more. TAFPred, employing an optimized Light Gradient Boosting Machine Regressor (LightGBM), achieved high accuracy with correlation coefficients of 0.746 and 0.737 and mean absolute errors of 0.114 and 0.123 for the φ and ψ angles, respectively. Compared to the state-of-the-art method, TAFPred demonstrated significant improvements of 10.08% in MAE and 24.83% in PCC for the phi angle and 9.93% in MAE, and 22.37% in PCC for the psi angle.

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/biology12071020

Cites

3 of 3 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Torsion angles phi (φ) and psi (ψ). The phi angle is the angle around the -N-CA- bond (where ‘CA’ is the alpha-carbon), and the psi angle is the angle around the -CA-C- bond.
Figure 1 Torsion angles phi (φ) and psi (ψ). The phi angle is the angle around the -N-CA- bond (where ‘CA’ is the alpha-carbon), and the psi angle is the angle around the -CA-C- bond.CC BY · PubMed Central
Figure 2 Illustration of the workflow of the torsion angle fluctuation predictions.
Figure 2 Illustration of the workflow of the torsion angle fluctuation predictions.CC BY · PubMed Central
Figure 3 Feature extraction from different tools.
Figure 3 Feature extraction from different tools.CC BY · PubMed Central
Figure 4 Selection of sliding window size with optimized LightGBM regressor (phi angle). Among the tested window sizes, it was found that a window size of 3 achieved the highest 1-MAE+PCC (mean absolute error + Pearson correlation coefficient) for the psi angle.
Figure 4 Selection of sliding window size with optimized LightGBM regressor (phi angle). Among the tested window sizes, it was found that a window size of 3 achieved the highest 1-MAE+PCC (mean absolute error + Pearson correlation coefficient) for the psi angle.CC BY · PubMed Central
Figure 5 Selection of sliding window size with optimized LightGBM regressor (psi angle). Among the tested window sizes, it was found that a window size of 3 yielded the highest value of 1-MAE+PCC for the psi angle.
Figure 5 Selection of sliding window size with optimized LightGBM regressor (psi angle). Among the tested window sizes, it was found that a window size of 3 yielded the highest value of 1-MAE+PCC for the psi angle.CC BY · PubMed Central
Figure 6 The torsion-angle fluctuation is depicted in its distribution, with the data points divided into 10 bins. The fluctuations of the phi and psi angles are visually represented using red and green colors, respectively.
Figure 6 The torsion-angle fluctuation is depicted in its distribution, with the data points divided into 10 bins. The fluctuations of the phi and psi angles are visually represented using red and green colors, respectively.CC BY · PubMed Central
Figure 7 The relationship between Δφ and Δψ is shown in the figure, revealing that the majority of residues exhibit small fluctuations below 0.2.
Figure 7 The relationship between Δφ and Δψ is shown in the figure, revealing that the majority of residues exhibit small fluctuations below 0.2.CC BY · PubMed Central
Figure 8 Relationship between torsion-angle fluctuation in the phi angle and disordered regions. The disordered probability was obtained from the SPOT-Disordered2 tool. The figure illustrates that regions with low disordered probability exhibit correspondingly low fluctuations in the phi angle, and conversely, regions with high disordered probability show higher fluctuations in the phi angle.
Figure 8 Relationship between torsion-angle fluctuation in the phi angle and disordered regions. The disordered probability was obtained from the SPOT-Disordered2 tool. The figure illustrates that regions with low disordered probability exhibit correspondingly low fluctuations in the phi angle, and conversely, regions with high disordered probability show higher fluctuations in the phi angle.CC BY · PubMed Central
Figure 9 Correlation between torsion-angle fluctuation in the psi angle and the presence of disordered regions. The disordered probability was obtained through the utilization of the SPOT-Disordered2 tool. The figure clearly illustrates that regions with low disordered probability exhibit lower fluctuations in the psi angle, while regions with high disordered probability tend to have higher fluctuations in the psi angle.
Figure 9 Correlation between torsion-angle fluctuation in the psi angle and the presence of disordered regions. The disordered probability was obtained through the utilization of the SPOT-Disordered2 tool. The figure clearly illustrates that regions with low disordered probability exhibit lower fluctuations in the psi angle, while regions with high disordered probability tend to have higher fluctuations in the psi angle.CC BY · PubMed Central

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.

A few free checks a day · papers already verified open instantly · results are public

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

Research works printed but not linked: 50

  1. Cornell, 1995, A second generation force field for the simulation of proteins, nucleic acids, and organic molecules, J. Am. Chem. Soc., 117, 5179, 10.1021/ja00124a002 · DOI
  2. Tompa, 2002, Intrinsically unstructured proteins, Trends Biol. Sci., 27, 527, 10.1016/S0968-0004(02)02169-2 · DOI
  3. Jonsson, 2017, Essential chemistry for biochemists, Essays Biochem., 61, 401, 10.1042/EBC20160094 · DOI
  4. Kulmanov, 2020, DeepGOPlus: Improved protein function prediction from sequence, Bioinformatics, 36, 422, 10.1093/bioinformatics/btz595 · DOI
  5. Nechab, 2014, 1,n-Hydrogen-Atom Transfer (HAT) Reactions in Which n≠ 5: An Updated Inventory, Chemistry, 20, 16034, 10.1002/chem.201403951 · DOI
  6. Wright, 1999, Intrinsically unstructured proteins: Re-assessing the protein structure-function paradigm, J. Mol. Biol., 293, 321, 10.1006/jmbi.1999.3110 · DOI · Paper page
  7. Quiocho, 1986, Carbohydrate-binding proteins: Tertiary structures and protein-sugar interactions, Annu. Rev. Biochem., 55, 287, 10.1146/annurev.bi.55.070186.001443 · DOI
  8. Mosimann, 1995, A critical assessment of comparative molecular modeling of tertiary structures of proteins, Proteins Struct. Funct. Bioinform., 23, 301, 10.1002/prot.340230305 · DOI
  9. Gao, J., Yang, Y., and Zhou, Y. (2018). Grid-based prediction of torsion angle probabilities of protein backbone and its application to discrimination of protein intrinsic disorder regions and selection of model structures. BMC Bioinform., 19. · DOI
  10. Heffernan, 2017, Capturing non-local interactions by long short-term memory bidirectional recurrent neural networks for improving prediction of protein secondary structure, backbone angles, contact numbers and solvent accessibility, Bioinformatics, 33, 2842, 10.1093/bioinformatics/btx218 · DOI
  11. Lyons, 2014, Predicting backbone Cα angles and dihedrals from protein sequences by stacked sparse auto-encoder deep neural network, J. Comput. Chem., 35, 2040, 10.1002/jcc.23718 · DOI
  12. Iqbal, S., and Hoque, M.T. (2015). DisPredict: A Predictor of Disordered Protein Using Optimized RBF Kernel. PLoS ONE, 10. · DOI
  13. Konrat, 2014, NMR contributions to structural dynamics studies of intrinsically disordered proteins, J. Magn. Reson., 241, 74, 10.1016/j.jmr.2013.11.011 · DOI
  14. Hu, 2021, flDPnn: Accurate intrinsic disorder prediction with putative propensities of disorder functions, Nat. Commun., 12, 4438, 10.1038/s41467-021-24773-7 · DOI
  15. Zhang, 2010, Fluctuations of backbone torsion angles obtained from NMR-determined structures and their prediction, Proteins, 78, 3353, 10.1002/prot.22842 · DOI
  16. Babu, 2011, Intrinsically disordered proteins: Regulation and disease, Curr. Opin. Struct. Biol., 21, 432, 10.1016/j.sbi.2011.03.011 · DOI
  17. Uversky, 2008, Intrinsically Disordered Proteins in Human Diseases: Introducing the D2 Concept, Annu. Rev. Biophys., 37, 215, 10.1146/annurev.biophys.37.032807.125924 · DOI
  18. Krishnan, 2022, De Novo Structure-Based Drug Design Using Deep Learning, J. Chem. Inf. Model., 62, 5100, 10.1021/acs.jcim.1c01319 · DOI
  19. Bulacu, 2013, Improved Angle Potentials for Coarse-Grained Molecular Dynamics Simulations, J. Chem. Theory Comput., 9, 3282, 10.1021/ct400219n · DOI
  20. Yee, 2005, NMR and X-ray Crystallography, Complementary Tools in Structural Proteomics of Small Proteins, J. Am. Chem. Soc., 127, 16512, 10.1021/ja053565+ · DOI
  21. Bryant, 1983, The NMR time scale, J. Chem. Educ., 60, 933, 10.1021/ed060p933 · DOI
  22. Schnapka, 2022, NMR Provides Unique Insight into the Functional Dynamics and Interactions of Intrinsically Disordered Proteins, Chem. Rev., 122, 9331, 10.1021/acs.chemrev.1c01023 · DOI
  23. Adamski, 2019, A Unified Description of Intrinsically Disordered Protein Dynamics under Physiological Conditions Using NMR Spectroscopy, J. Am. Chem. Soc., 141, 17817, 10.1021/jacs.9b09002 · DOI
  24. Kosol, 2013, Structural characterization of intrinsically disordered proteins by NMR spectroscopy, Molecules, 18, 10802, 10.3390/molecules180910802 · DOI
  25. Graether, 2019, Troubleshooting Guide to Expressing Intrinsically Disordered Proteins for Use in NMR Experiments, Front. Mol. Biosci., 5, 118, 10.3389/fmolb.2018.00118 · DOI
  26. Yang, 2011, Improving protein fold recognition and template-based modeling by employing probabilistic-based matching between predicted one-dimensional structural properties of query and corresponding native properties of templates, Bioinformatics, 27, 2076, 10.1093/bioinformatics/btr350 · DOI
  27. Karchin, 2003, Hidden Markov models that use predicted local structure for fold recognition: Alphabets of backbone geometry, Proteins Struct. Funct. Bioinform., 51, 504, 10.1002/prot.10369 · DOI
  28. Rohl, C.A., Strauss, C.E., Misura, K.M., and Baker, D. (2004). Methods in Enzymology, Elsevier.
  29. Faraggi, 2009, Predicting continuous local structure and the effect of its substitution for secondary structure in fragment-free protein structure prediction, Structure, 17, 1515, 10.1016/j.str.2009.09.006 · DOI
  30. Wu, S., and and Zhang, Y. (2008). ANGLOR: A composite machine-learning algorithm for protein backbone torsion angle prediction. PLoS ONE, 3. · DOI
  31. Yang, 2018, Sixty-five years of the long march in protein secondary structure prediction: The final stretch?, Brief. Bioinform., 19, 482
  32. Li, H., Hou, J., Adhikari, B., Lyu, Q., and Cheng, J. (2017). Deep learning methods for protein torsion angle prediction. BMC Bioinform., 18. · DOI
  33. Wu, R., Ding, F., Wang, R., Shen, R., Zhang, X., Luo, S., Su, C., Wu, Z., Xie, Q., and Berger, B. (2022). High-resolution de novo structure prediction from primary sequence. bioRxiv. · DOI · Paper page
  34. Lin, Z., Akin, H., Rao, R., Hie, B., Zhu, Z., Lu, W., dos Santos Costa, A., Fazel-Zarandi, M., Sercu, T., and Candido, S. (2022). Language models of protein sequences at the scale of evolution enable accurate structure prediction. bioRxiv.
  35. Kabir, M.W.U., Alawad, D.M., Mishra, A., and Hoque, M.T. (2023, January 29–31). Prediction of Phi and Psi Angle Fluctuations from Protein Sequences. Proceedings of the 20th IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, Eindhoven, The Netherlands. · DOI · Paper page
  36. Iqbal, 2015, Improved Prediction of Accessible Surface Area Results in Efficient Energy Function Application, J. Theor. Biol., 380, 380, 10.1016/j.jtbi.2015.06.012 · DOI
  37. Iqbal, 2018, PBRpredict-Suite: A suite of models to predict peptide-recognition domain residues from protein sequence, Bioinformatics, 34, 3289, 10.1093/bioinformatics/bty352 · DOI
  38. Iqbal, S., and Hoque, M.T. (2016). Estimation of position specific energy as a feature of protein residues from sequence alone for structural classification. PLoS ONE, 11. · DOI
  39. Zhu, 2010, Improving the accuracy of predicting disulfide connectivity by feature selection, Comput. Chem., 31, 1478, 10.1002/jcc.21433 · DOI
  40. Islam, 2016, A balanced secondary structure predictor, J. Theor. Biol., 389, 60, 10.1016/j.jtbi.2015.10.015 · DOI
  41. Liu, 2002, Loopy proteins appear conserved in evolution, J. Mol. Biol., 322, 53, 10.1016/S0022-2836(02)00736-2 · DOI
  42. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., and Liu, T.-Y. (2017, January 4–9). LightGBM: A highly efficient gradient boosting decision tree. Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA.
  43. Chen, T., and Guestrin, C. (2016, January 13–17). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (ACM, 2016), San Francisco, CA, USA. · DOI · Paper page
  44. Geurts, 2006, Extremely randomized trees, Mach. Learn., 63, 3, 10.1007/s10994-006-6226-1 · DOI · Paper page
  45. Hastie, T., Tibshirani, R., and Friedman, J. (2009). The Elements of Statistical Learning, Springer. [2nd ed.]. · DOI · Paper page
  46. Altman, 1992, An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression, Am. Stat., 46, 175, 10.1080/00031305.1992.10475879 · DOI
  47. Arik, S.O., and Pfister, T. (2019). TabNet: Attentive Interpretable Tabular Learning. arXiv.
  48. Hoque, 2017, Genetic algorithm-based improved sampling for protein structure prediction, Int. J. Bio-Inspired Comput., 9, 129, 10.1504/IJBIC.2017.083702 · DOI
  49. Hoque, M.T., Chetty, M., and Sattar, A. (2007, January 25–28). Protein Folding Prediction in 3D FCC HP Lattice Model using Genetic Algorithm. Proceedings of the IEEE Congress on Evolutionary Computation (CEC), Singapore. · DOI
  50. Hoque, 2010, DFS Generated Pathways in GA Crossover for Protein Structure Prediction, Neurocomputing, 73, 2308, 10.1016/j.neucom.2010.02.021 · DOI

Metadata

Authors
Kabir, M. W. U., Alawad, D. M., Mishra, A., & Hoque, M. T.
Affiliations
  • LSU New Orleans (US)
  • Texas A&M University – Kingsville (US)
Corresponding author
Hoque, Md Tamjidul
ORCID
Venue
Biology
Publisher
MDPI AG
Published
2023-07-19
Volume
12
Issue
7
Pages
1020
ISSN
2079-7737 (ISSN-L)
DOI
10.3390/biology12071020
PMC
PMC10376102
OpenAlex
W4384929351
Cited by
5
References
3
Citations by year
2026: 1 · 2025: 1 · 2024: 2 · 2023: 1
FWCI
0.45
Type
journal article
Language
en
Open access
Licence
Locations
  • Biology (publishedVersion, cc-by)
  • Preprints.org (acceptedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Department of Homeland Security (DHS) (21STSLA00011-01-0)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Structural Biology › Enzyme Structure and Function
Keywords
phi and psi angles, protein flexibility, sequence-based prediction, LightGBM regression

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