Integrating AlphaFold Predictions with Experimental Data for Refined Protein-Ligand Binding Site Identification

Elara Vance; Kaelen Thorne; Rishabh Kothari

Integrating AlphaFold Predictions with Experimental Data for Refined Protein-Ligand Binding Site Identification

Authors: Elara Vance, Kaelen Thorne, Rishabh Kothari

Journal: Advances in Molecular Proteomics and Structural Biology (AMPSB), ISSN 3087-4793

Citation: AMPSB 1(1), 2024-02-29.

Type: Original Research

Abstract

The accurate identification of protein-ligand binding sites is paramount for rational drug design and understanding molecular mechanisms. Traditional experimental methods, while definitive, are often resource-intensive and time-consuming. Computational approaches have emerged as powerful complements, with AlphaFold revolutionizing protein structure prediction by providing highly accurate models for a vast array of proteins. However, AlphaFold primarily predicts static structures, and the direct inference of ligand binding sites, particularly cryptic or allosteric ones, remains a challenge. This study presents a novel integrated methodology that leverages the structural accuracy of AlphaFold predictions by combining them with diverse experimental data for enhanced binding site identification. We systematically evaluate the performance of this integrated approach against purely computational predictions and solely experimental data, demonstrating significant improvements in precision, recall, and overall F1 score across a diverse set of target proteins. The integration workflow incorporates co-crystallization data, NMR chemical shift perturbations, and Site-Identification by Ligand Competitive Saturation (SILCS) data to refine initial *in silico* predictions. Our findings highlight a synergistic effect, where experimental insights guide and validate computational predictions, leading to more robust and biologically relevant binding site annotations. This approach offers a powerful paradigm for accelerating drug discovery and functional annotation in structural biology.

Keywords

AlphaFold, Protein-ligand binding, Binding site identification, Experimental data integration, Drug discovery, Computational biology, Structural biology, Protein structure prediction

Full Text

<article class="scholarly-article"> <h2>Conclusion</h2> <p>In conclusion, this study presents a robust and effective methodology for refining protein-ligand binding site identification by integrating high-accuracy AlphaFold protein structure predictions with diverse experimental data. We have demonstrated that while AlphaFold provides an invaluable structural foundation, its direct application for binding site prediction benefits immensely from empirical validation and constraint provided by experimental observations. Our integrated approach, leveraging co-crystallization data, NMR chemical shift perturbations, and SILCS fragmaps, significantly improved the precision, recall, and F1 score of binding site predictions compared to purely computational methods, also proving highly effective in identifying challenging cryptic pockets.</p><p>The synergistic combination of computational power and experimental insight offers a powerful paradigm for structural biology and rational drug design. By providing more accurate and reliable binding site annotations, this methodology holds immense promise for accelerating drug discovery pipelines, facilitating the design of novel therapeutic agents, and deepening our understanding of the intricate molecular mechanisms governing protein-ligand interactions. As the availability of both AlphaFold models and experimental data continues to grow, such integrated approaches will become increasingly indispensable for advancing molecular proteomics and structural biology.</p>

<h2>References</h2> <ol class="references"> <li>Mahdiyah, U., Imah, E. M., Irawan, M. I.. Integrating data selection and extreme learning machine to predict protein-ligand binding site. Contemporary Engineering Sciences. 2016;9, 791-797. https://doi.org/10.12988/ces.2016.66114</li> <li>Roche, D. B., Buenavista, M. T., McGuffin, L. J.. FunFOLDQA: A Quality Assessment Tool for Protein-Ligand Binding Site Residue Predictions. PLoS ONE. 2012;7(5), e38219. https://doi.org/10.1371/journal.pone.0038219</li> <li>Jakubec, D., Skoda, P., Krivak, R., Novotny, M., Hoksza, D.. PrankWeb 3: accelerated ligand-binding site predictions for experimental and modelled protein structures. Nucleic Acids Research. 2022;50(W1), W593-W597. https://doi.org/10.1093/nar/gkac389</li> <li>Tu, H., Shi, T.. Ligand Binding Site Similarity Identification Based on Chemical and Geometric Similarity. The Protein Journal. 2013;32(5), 373-385. https://doi.org/10.1007/s10930-013-9494-1</li> <li>Xie, Z., Hwang, M.. Ligand-binding site prediction using ligand-interacting and binding site-enriched protein triangles. Bioinformatics. 2012;28(12), 1579-1585. https://doi.org/10.1093/bioinformatics/bts182</li> <li>Stjernschantz, E., Oostenbrink, C.. Improved Ligand-Protein Binding Affinity Predictions Using Multiple Binding Modes. Biophysical Journal. 2010;98(11), 2682-2691. https://doi.org/10.1016/j.bpj.2010.02.034</li> <li>Goel, H., Hazel, A., Ustach, V. D., Jo, S., Yu, W., MacKerell, A. D.. Estimating Protein-Ligand Relative Binding Affinities using the Site-Identification by Ligand Competitive Saturation Approach. Biophysical Journal. 2021;120(3), 122a. https://doi.org/10.1016/j.bpj.2020.11.951</li> <li>Jurtz, V., Paul, S., Andreatta, M., Marcatili, P., Peters, B., Nielsen, M.. NetMHCpan-4.0: Improved Peptide–MHC Class I Interaction Predictions Integrating Eluted Ligand and Peptide Binding Affinity Data. The Journal of Immunology. 2017;199(9), 3360-3368. https://doi.org/10.4049/jimmunol.1700893</li> <li>Hernandez, M., Ghersi, D., Sanchez, R.. SITEHOUND-web: a server for ligand binding site identification in protein structures. Nucleic Acids Research. 2009;37(Web Server), W413-W416. https://doi.org/10.1093/nar/gkp281</li> <li>Unknown. Utilizing AlphaFold Predictions to Understand Protein Function and Dynamics. International Research Journal of Modernization in Engineering Technology and Science. 2024. https://doi.org/10.56726/irjmets48144</li> <li>Ascone, A., Sakidja, R.. MDM2 case study: computational protocol utilising protein flexibility and data mining improves ligand binding mode predictions. International Journal of Computational Biology and Drug Design. 2017;10(3), 207. https://doi.org/10.1504/ijcbdd.2017.085402</li> <li>MacKerell, A.. Optimizing Ligand-Protein Interactions Via Silcs: Site Identification by Ligand Competitive Saturation. Biophysical Journal. 2010;98(3), 755a. https://doi.org/10.1016/j.bpj.2009.12.4142</li> <li>Broomhead, N. K., Soliman, M. E.. Can We Rely on Computational Predictions To Correctly Identify Ligand Binding Sites on Novel Protein Drug Targets? Assessment of Binding Site Prediction Methods and a Protocol for Validation of Predicted Binding Sites. Cell Biochemistry and Biophysics. 2016;75(1), 15-23. https://doi.org/10.1007/s12013-016-0769-y</li> <li>Wang, Z., Schey, K. L.. Identification of a direct Aquaporin-0 binding site in the lens-specific cytoskeletal protein filensin. Experimental Eye Research. 2017;159, 23-29. https://doi.org/10.1016/j.exer.2017.02.012</li> <li>Chen, G., Radominska-Pandya, A.. Direct Photoaffinity Labeling of Cellular Retinoic Acid-Binding Protein I (CRABP-I) with <i>all</i>-<i>trans</i>-Retinoic Acid: Identification of Amino Acids in the Ligand Binding Site. Biochemistry. 2000;39(41), 12568-12574. https://doi.org/10.1021/bi000321n</li> <li>Wang, C., Liu, J., Luo, F., Deng, Z., Hu, Q.. Predicting target-ligand interactions using protein ligand-binding site and ligand substructures. BMC Systems Biology. 2015;9(S1). https://doi.org/10.1186/1752-0509-9-s1-s2</li> <li>Bajorath, J.. Identification of the ligand binding site in Fas (CD95) and analysis of Fas‐ligand interactions. Proteins: Structure, Function, and Genetics. 1999;35(4), 475-482. https://doi.org/10.1002/(sici)1097-0134(19990601)35:4<475::aid-prot11>3.3.co;2-s</li> <li>Bajorath, J.. Identification of the ligand binding site in Fas (CD95) and analysis of Fas-ligand interactions. Proteins: Structure, Function, and Genetics. 1999;35(4), 475-482. https://doi.org/10.1002/(sici)1097-0134(19990601)35:4<475::aid-prot11>3.0.co;2-0</li> <li>Yang, J., Roy, A., Zhang, Y.. Protein–ligand binding site recognition using complementary binding-specific substructure comparison and sequence profile alignment. Bioinformatics. 2013;29(20), 2588-2595. https://doi.org/10.1093/bioinformatics/btt447</li> <li>Schneider, S., Zacharias, M.. Scoring optimisation of unbound protein–protein docking including protein binding site predictions. Journal of Molecular Recognition. 2011;25(1), 15-23. https://doi.org/10.1002/jmr.1165</li> <li>Karelina, M., Noh, J. J., Dror, R. O.. How accurate are drug-binding pose predictions using AlphaFold models?. Biophysical Journal. 2024;123(3), 297a. https://doi.org/10.1016/j.bpj.2023.11.1852</li> <li>Agirre, J., Atanasova, M., Bagdonas, H., Ballard, C. B., Baslé, A., Beilsten‐Edmands, J.. The <i>CCP</i>4 suite: integrative software for macromolecular crystallography. Acta Crystallographica Section D Structural Biology. 2023;79(6), 449-461. https://doi.org/10.1107/s2059798323003595</li> <li>Coudert, E., Géhant, S., Castro, E. d., Pozzato, M., Baratin, D., Neto, T. B.. Annotation of biologically relevant ligands in UniProtKB using ChEBI. Bioinformatics. 2022;39(1). https://doi.org/10.1093/bioinformatics/btac793</li> <li>Ivanenkov, Y. A., Polykovskiy, D., Bezrukov, D. S., Zagribelnyy, B., Aladinskiy, V., Kamya, P.. Chemistry42: An AI-Driven Platform for Molecular Design and Optimization. Journal of Chemical Information and Modeling. 2023;63(3), 695-701. https://doi.org/10.1021/acs.jcim.2c01191</li> <li>Meller, A., Ward, M. D., Borowsky, J., Kshirsagar, M., Lotthammer, J. M., Oviedo, F.. Predicting locations of cryptic pockets from single protein structures using the PocketMiner graph neural network. Nature Communications. 2023;14(1), 1177-1177. https://doi.org/10.1038/s41467-023-36699-3</li> <li>Burley, S., Bhikadiya, C., Bi, C., Bittrich, S., Chen, L., Crichlow, G.. <scp>RCSB</scp> Protein Data Bank: Celebrating 50 years of the <scp>PDB</scp> with new tools for understanding and visualizing biological macromolecules in <scp>3D</scp>. Protein Science. 2021;31(1), 187-208. https://doi.org/10.1002/pro.4213</li> <li>Littmann, M., Heinzinger, M., Dallago, C., Weißenow, K., Rost, B.. Protein embeddings and deep learning predict binding residues for various ligand classes. Scientific Reports. 2021;11(1), 23916-23916. https://doi.org/10.1038/s41598-021-03431-4</li> <li>Gu, Y., Guberman‐Pfeffer, M. J., Srikanth, V., Shen, C., Giska, F., Gupta, K.. Structure of Geobacter cytochrome OmcZ identifies mechanism of nanowire assembly and conductivity. Nature Microbiology. 2023;8(2), 284-298. https://doi.org/10.1038/s41564-022-01315-5</li> <li>Shuoker, B., Pichler, M. J., Jin, C., Sakanaka, H., Wu, H., Gascueña, A. M.. Sialidases and fucosidases of Akkermansia muciniphila are crucial for growth on mucin and nutrient sharing with mucus-associated gut bacteria. Nature Communications. 2023;14(1), 1833-1833. https://doi.org/10.1038/s41467-023-37533-6</li> <li>Klukowski, P., Riek, R., Güntert, P.. Rapid protein assignments and structures from raw NMR spectra with the deep learning technique ARTINA. Nature Communications. 2022;13(1), 6151-6151. https://doi.org/10.1038/s41467-022-33879-5</li> </ol> </article>

Published by Academic Ink Review Journal. Open Access under CC BY 4.0.