Abstract

High cost of HTS and lengthy development time are two traditional drawbacks to the discovery of new peptide therapeutics. The authors of this study suggest an integrated computational framework that integrates Large Language Models (LLMs) and Scientific Knowledge Graphs (SKG) to speed up the identification of bioactive peptides. We took advantage of a specific dataset of 432 examples of peptide-protein interactions to build a generative model that was fine-tuned based on the sequence-structure characteristics and a biological pathway knowledge graph. The LLM serves as a model for generating peptide sequences, and the knowledge graph makes sure that the candidate compounds are biologically relevant. Our results show that this hybrid approach can increase the accuracy of the prediction of the binding affinity of the models individually. The study relies on data integration and modelling with Python, PyTorch and Neo4j. The framework can be used to effectively constrain sequence space by imposing structural constraints, thereby limiting the number of experiments needed to identify strong therapeutic candidates. This strategy has the potential to be a blueprint for peptide drug research and shows that structured knowledge can be combined with the latent representations of language to create a strong pipeline for fast peptide design for clinical targets.

References

• D. Paul, G. Sanap, S. Shenoy, D. Kalyane, K. Kalia, and R. K. Tekade, “Artificial intelligence in drug discovery and development,” Drug Discovery Today, vol. 26, no. 1, pp. 80–93, 2021. DOI: https://doi.org/10.1016/j.drudis.2020.10.010
• K. K. Mak and M. R. Pichika, “Artificial intelligence in drug development: Present status and future prospects,” Drug Discovery Today, vol. 24, no. 3, pp. 773–780, 2019. DOI: https://doi.org/10.1016/j.drudis.2018.11.014
• S. Priya, G. Tripathi, D. B. Singh, P. Jain, and A. Kumar, “Machine learning approaches and their applications in drug discovery and design,” Chemical Biology & Drug Design, vol. 100, no. 1, pp. 136–153, 2022. DOI: https://doi.org/10.1111/cbdd.14057
• R. Inaganti, “Navigating FDA regulations for combination products: Challenges at the intersection of drugs and devices,” FMDB Transactions on Sustainable Health Science Letters, vol. 2, no. 3, pp. 164–174, 2024. DOI: 10.69888/FTSHSL.2024.000252
• I. Lee, J. Keum, and H. Nam, “DeepConv-DTI: Prediction of drug-target interactions via deep learning with convolution on protein sequences,” PLoS Computational Biology, vol. 15, no. 6, p. e1007129, 2019. DOI: https://doi.org/10.1371/journal.pcbi.1007129
• K. Huang, T. Fu, L. M. Glass, M. Zitnik, C. Xiao, and J. Sun, “DeepPurpose: A deep learning library for drug-target interaction prediction,” Bioinformatics, vol. 36, nos. 22–23, pp. 5545–5547, 2021. DOI: https://doi.org/10.1093/bioinformatics/btaa1005
• B. Sanchez-Lengeling, C. Outeiral, G. L. Guimaraes, and A. Aspuru-Guzik, “Optimizing distributions over molecular space: An objective-reinforced generative adversarial network for inverse-design chemistry (ORGANIC),” 2017. DOI: https://doi.org/10.26434/chemrxiv.5309668.v3
• R. Gomez-Bombarelli, J. N. Wei, D. Duvenaud, J. M. Hernandez-Lobato, B. Sanchez-Lengeling, D. Sheberla, T. D. Hirzel, R. P. Adams, and A. Aspuru-Guzik, “Automatic chemical design using a data-driven continuous representation of molecules,” ACS Central Science, vol. 4, no. 2, pp. 268–276, 2018. DOI: https://doi.org/10.1021/acscentsci.7b00572
• H. Askr, E. Elgeldawi, H. Aboul Ella, Y. Elshaier, M. M. Gomaa, and A. E. Hassanien, “Deep learning in drug discovery: An integrative review and future challenges,” Artificial Intelligence Review, vol. 56, no. 7, pp. 5975–6037, 2023. DOI: https://doi.org/10.1007/s10462-022-10306-1
• A. Blanco-Gonzalez, A. Cabezon, A. Seco-Gonzalez, D. Conde-Torres, P. Antelo-Riveiro, A. Pineiro, and R. Garcia-Fandino, “The role of AI in drug discovery: Challenges, opportunities, and strategies,” Pharmaceuticals, vol. 16, no. 6, 2023. DOI: https://doi.org/10.3390/ph16060891
• R. Qureshi, M. Irfan, T. M. Gondal, S. Khan, J. Wu, M. U. Hadi, J. Heymach, L. X. Le, H. Yan, and T. Alam, “AI in drug discovery and its clinical relevance,” Heliyon, vol. 9, no. 7, p. e17575, 2023. DOI: https://doi.org/10.1016/j.heliyon.2023.e17575
• A. V. Sadybekov and V. Katritch, “Computational approaches streamlining drug discovery,” Nature, vol. 616, no. 7958, pp. 673–685, 2023. DOI: https://doi.org/10.1038/s41586-023-05905-z
• Y. A. Ivanenkov, D. Polykovskiy, D. Bezrukov, B. Zagribelnyy, V. Aladinskiy, P. Kamya, A. Aliper, F. Ren, and A. Zhavoronkov, “Chemistry42: An AI-driven platform for molecular design and optimization,” Journal of Chemical Information and Modeling, vol. 63, no. 3, pp. 695–701, 2023.
• P. Kamya, I. V. Ozerov, F. W. Pun, K. Tretina, T. Fokina, S. Chen, V. Naumov, X. Long, S. Lin, and M. Korzinkin, “PandaOmics: An AI-driven platform for therapeutic target and biomarker discovery,” Journal of Chemical Information and Modeling, vol. 64, no. 10, pp. 3961–3969, 2024.
• A. Gangwal and A. Lavecchia, “AI-driven drug discovery for rare diseases,” Journal of Chemical Information and Modeling, 2024. DOI: https://doi.org/10.1021/acs.jcim.4c01966