Related Experiment Video
Updated: Jul 30, 2025
Predictive Immune Modeling of Solid Tumors
Published on: February 25, 2020
In Silico Pipeline to Identify Tumor-Specific Antigens for Cancer Immunotherapy Using Exome Sequencing Data
Diego Morazán-Fernández1, Javier Mora2, Jose Arturo Molina-Mora2
1Caja Costarricense de Seguro Social, San José, 10104 Costa Rica.
Abstract:
Tumor-specific antigens or neoantigens are peptides that are expressed only in cancer cells and not in healthy cells. Some of these molecules can induce an immune response, and therefore, their use in immunotherapeutic strategies based on cancer vaccines has been extensively explored. Studies based on these approaches have been triggered by the current high-throughput DNA sequencing technologies. However, there is no universal nor straightforward bioinformatic protocol to discover neoantigens using DNA sequencing data. Thus, we propose a bioinformatic protocol to detect tumor-specific antigens associated with single nucleotide variants (SNVs) or "mutations" in tumoral tissues. For this purpose, we used publicly available data to build our model, including exome sequencing data from colorectal cancer and healthy cells obtained from a single case, as well as frequent human leukocyte antigen (HLA) class I alleles in a specific population. HLA data from Costa Rican Central Valley population was selected as an example. The strategy included three main steps: (1) pre-processing of sequencing data; (2) variant calling analysis to detect tumor-specific SNVs in comparison with healthy tissue; and (3) prediction and characterization of peptides (protein fragments, the tumor-specific antigens) derived from the variants, in the context of their affinity with frequent alleles of the selected population. In our model data, we found 28 non-silent SNVs, present in 17 genes in chromosome one. The protocol yielded 23 strong binders peptides derived from the SNVs for frequent HLA class I alleles for the Costa Rican population. Although the analyses were performed as an example to implement the pipeline, to our knowledge, this is the first study of an in silico cancer vaccine using DNA sequencing data in the context of the HLA alleles. It is concluded that the standardized protocol was not only able to identify neoantigens in a specific but also provides a complete pipeline for the eventual design of cancer vaccines using the best bioinformatic practices.
Supplementary Information:
The online version contains supplementary material available at 10.1007/s43657-022-00084-9.
Insights
This study introduces a bioinformatic protocol to identify tumor-specific antigens (neoantigens) from DNA sequencing data, aiding in cancer vaccine development. The method successfully pinpointed potential neoantigens for a specific population, advancing personalized cancer immunotherapy.
Area of Science:
- Oncology
- Bioinformatics
- Immunology
Background:
- Tumor-specific antigens (neoantigens) are crucial targets for cancer immunotherapy and vaccines.
- High-throughput DNA sequencing technologies have enabled neoantigen discovery, but standardized bioinformatic protocols are lacking.
- Identifying neoantigens requires analyzing tumor DNA for mutations and predicting their immunogenic potential.
Purpose of the Study:
- To develop and validate a straightforward bioinformatic protocol for discovering tumor-specific antigens (neoantigens) from DNA sequencing data.
- To identify neoantigens associated with single nucleotide variants (SNVs) in tumor tissues.
- To provide a pipeline for the in silico design of personalized cancer vaccines.
Main Methods:
- A three-step bioinformatic protocol was established: data pre-processing, variant calling for tumor-specific SNVs, and neoantigen prediction.
- Publicly available exome sequencing data from colorectal cancer and healthy cells were utilized.
- Human Leukocyte Antigen (HLA) class I alleles specific to the Costa Rican Central Valley population were incorporated for peptide binding predictions.
Main Results:
- The protocol identified 28 non-silent SNVs across 17 genes in chromosome one.
- A total of 23 high-affinity binding peptides, predicted as neoantigens, were identified for frequent HLA class I alleles in the selected population.
- This represents a novel in silico approach for cancer vaccine design using DNA sequencing and HLA data.
Conclusions:
- The developed standardized bioinformatic protocol effectively identifies neoantigens from SNVs.
- This pipeline offers a comprehensive approach for the in silico design of cancer vaccines.
- The study highlights the potential of bioinformatic tools in advancing personalized cancer immunotherapy.
