Related Experiment Video
Updated: Jun 16, 2026
06:32
Multiplex Immunohistochemical Analysis of the Spatial Immune Cell Landscape of the Tumor Microenvironment
Published on: August 18, 2023
A novel computational framework for tumor-specific T cell antigen identification using a deep neural network
Salman Khan1, Islam Uddin2, Fawaz Khaled Alarfaj3
1Department of Computer Science, College of Computer and Information Sciences, King Saud University, 11451, Riyadh, Saudi Arabia.
Journal of Computer-Aided Molecular Design
|June 15, 2026
Summary
This study introduces a novel Deep Neural Network framework for identifying tumor-specific T-cell antigens, improving cancer immunotherapy precision. The AI-driven approach enhances antigen prediction accuracy and efficiency, outperforming existing methods.
Area of Science:
- Oncology
- Immunology
- Bioinformatics
Background:
- Accurate identification of tumor-specific T-cell antigens is critical for advancing cancer immunotherapy.
- Current artificial intelligence (AI) and machine learning (ML) methods face challenges with the complexity and sequence dependency of antigen data, leading to suboptimal predictions.
Purpose of the Study:
- To develop a Deep Neural Network (DNN)-based framework for enhanced computational tumor T-cell antigen identification.
- To address limitations in existing AI/ML approaches for antigen prediction.
Main Methods:
- Utilized hybrid sequence encoding: Position-Specific Scoring Matrix with Discrete Wavelet Transform (PsePSSM-DWT) and Protein Bidirectional Encoder Representations from Transformers (ProtBERT-BFD).
- Implemented a Shapley Additive exPlanations (SHAP)-based global feature selection strategy for efficient feature selection.
- Trained a DNN model on the optimized feature set for antigen identification.
Main Results:
- The proposed DNN framework achieved an average accuracy of 96.16% and a Matthew's correlation coefficient of 0.923.
- Demonstrated significant performance improvement over conventional machine learning and state-of-the-art methods.
- Established a robust computational baseline for T-cell antigen identification.
Conclusions:
- The developed framework offers a powerful tool for precision-driven, AI-assisted discovery of tumor antigens.
- Provides a foundation for integrating multi-omics data and real-time immunotherapy workflows.
- Represents a significant advancement in computational approaches for cancer immunotherapy.
