Related Experiment Video
Updated: Jul 12, 2026
Predictive Immune Modeling of Solid Tumors
Published on: February 25, 2020
Enhancing tumor T cell antigen prediction by integrating deep protein representations
Umut Oskay1, Baris Tudes1, Emre Sefer2
1Artificial Intelligence and Data Engineering Department, Ozyegin University, Orman Sokak, 34794, Cekmekoy, Istanbul, Türkiye.
Background:
Cancer is a major threat to human health, and cancer immunotherapy is considered the most promising treatment option. It offers high effectiveness and precision, with fewer side effects than traditional treatments. Tumor T cell antigens (TTCA) are the proteins or protein fragments that stand on the surface of cancer cells and that are recognized by the immune system. However, physical experimental approaches to infer them might be costly and time-consuming, even though they are important for cancer immunotherapy.
Results:
We propose DEEPTHYBRID to predict TTCAs by combining chemically hand-designed features with deep protein representations from ProtBERT, an adapted version of the BERT large Language Model, for finding complex representations from the protein sequences. It is the first framework to incorporate these protein embeddings with complementary chemical features for this problem. After the feature extraction step, among multiple classifiers tested, we find Inception-based neural network to outperform the rest, so DEEPTHYBRID includes it. We evaluate the prediction performance on multiple datasets, where DEEPTHYBRID consistently outperforms the competing approaches across all datasets. For instance, on the first dataset, DEEPTHYBRID achieves an accuracy of 0.76, an F1-score of 0.78, outperforming existing machine learning and state-of-the-art TTCA predictors. Across both datasets, combining deep protein embeddings with chemical attributes yields superior performance compared to using either feature type alone. We also find the mixture of deep features and chemical features more informative in terms of Shapley-based explanation values.
Conclusions:
Overall, our method is promising with its high accuracy rates and strong predictive skills. In addition to its predictive performance, DEEPTHYBRID provides a fast and scalable computational screening framework that can substantially reduce the cost and time associated with biological experiments by prioritizing high-confidence tumor T cell antigen candidates for downstream experimental validation.
Insights
DEEPTHYBRID predicts tumor T cell antigens (TTCA) using deep protein features and chemical attributes. This computational approach accelerates cancer immunotherapy research by reducing experimental costs and time.
Area of Science:
- Immunology
- Computational Biology
- Bioinformatics
Background:
- Cancer immunotherapy is a promising treatment with high effectiveness and fewer side effects.
- Tumor T cell antigens (TTCA) are crucial targets for immunotherapy but are costly and time-consuming to identify experimentally.
- Developing efficient methods to predict TTCA is essential for advancing cancer treatment.
Purpose of the Study:
- To develop a novel computational framework, DEEPTHYBRID, for accurate prediction of tumor T cell antigens (TTCA).
- To integrate deep protein representations with chemical features for enhanced TTCA prediction.
- To provide a fast and scalable alternative to experimental methods for identifying TTCA candidates.
Main Methods:
- Utilized ProtBERT, a BERT-based language model, to extract deep protein representations from amino acid sequences.
- Incorporated chemically hand-designed features alongside deep protein embeddings.
- Employed an Inception-based neural network as the primary classifier within the DEEPTHYBRID framework.
Main Results:
- DEEPTHYBRID demonstrated superior performance across multiple datasets, outperforming existing TTCA predictors.
- Achieved an accuracy of 0.76 and an F1-score of 0.78 on one dataset, surpassing state-of-the-art methods.
- Combining deep protein embeddings with chemical attributes proved more effective than using either feature type alone.
Conclusions:
- DEEPTHYBRID offers a highly accurate and efficient computational method for predicting tumor T cell antigens.
- The framework significantly reduces the cost and time required for experimental validation of TTCA candidates.
- DEEPTHYBRID accelerates the discovery of potential targets for cancer immunotherapy development.
