Loading...

MHC-Peptide Binding Prediction Using a Deep Learning Method with Efficient GPU Implementation Approach

Darvishi, Saeed | 2023

61 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56296 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Koohi, Somayyeh
  7. Abstract:
  8. The Major Histocompatibility Complex (MHC) binds to the derived peptides from pathogens to present them to killer T cells on the cell surface. Developing computational methods for accurate, fast, and explainable peptide-MHC binding prediction can facilitate immunotherapies and vaccine development. Various deep learning-based methods rely on feature extraction from the peptide and MHC sequences separately and ignore their valuable binding information. This paper develops a capsule neural network-based method to efficiently capture and model the peptide-MHC complex features to predict the peptide- MHC class I binding. Various evaluations over multiple datasets using popular performance metrics showed that our method outperformed baseline and state-of-the-art methods for predicting peptide binding to the MHC, while it is able to provide accurate prediction over less available binding data. Moreover, for providing precise insights into the results, we explored the essential features contributed to the prediction taking advantages of feature importance algorithms. Since the simulation results demonstrated consistency with the experimental studies, we concluded that our method can be utilized for the accurate, rapid, and interpretable peptide-MHC binding prediction to assist biological therapies
  9. Keywords:
  10. Deep Learning ; Peptides ; Major Histocompatibility Complex (MHC) ; Capsule Deep Neural Network ; Graphics Procssing Unit (GPU) ; Binding Affinity

 Digital Object List

 Bookmark

No TOC