In silico neoantigen prediction in glioblastoma for personalised immunotherapy

Glioblastoma multiforme is a type IV glioma and is one of the most fatal cancers affecting the Brain and the Spinal cord. Even with medications, patients only survive for about 15 months.  Therefore, personalised immunotherapy has become an avid area of research. Neoantigen based vaccines are gaining further recognition because herein the immune system specifically identifies cancer cells and not the healthy ones. Neoantigens are peptides that form within cancer cells when they undergo somatic mutations. These neoantigens then bind to an HLA molecule causing them to be picked up by a T cell which then triggers an immune response. 
In this article I showcase a bioinformatic pipeline to identify peptides that can be further analysed for personalised immunotherapy in glioblastoma patients. A publicly available RNA sequencing data from NCBI SRA database was used for this project. The pipeline includes quality control, read alignment, variant calling, annotation, HLA typing and binding prediction. The neoantigen binding predictions were carried out using HLA-A*02:01.
Out of all the peptides analysed, VSDPGQLEHV showed the strongest binding to HLA-A*02:01 and a percentile rank of 2.4. Overall, this article gives a workflow that can be followed to identify neoantigens and therefore showcases the importance of computational analysis in personalised immunotherapeutics.