Neoantigen Prediction Pipeline

A bioinformatics pipeline for neoantigen prediction from tumour samples, developed as an MSc Bioinformatics thesis at Cardiff University.

Live thesis: https://mylonas.github.io/neoantigen

Overview

The project presents evidence for the importance of neoantigens in tumour samples and reviews pipelines for their prediction. Neoantigens are tumour-specific mutant peptides that can be recognised by the immune system, making them key targets for personalised cancer immunotherapy.

Contents

File	Description
thesis.pdf	MSc thesis: neoantigen prediction pipeline review and analysis
rna_seq.py	Python script for RNA-seq data processing
00_workflow.sh	Shell script orchestrating the full prediction pipeline
index.html	GitHub Pages PDF viewer

Dependencies

Tool	Purpose	Source
Opossum	Variant filtering	BSGOxford/OpossumDependencies
Platypus	Variant calling	andyrimmer/Platypus
ArcasHLA	HLA typing from RNA-seq	RabadanLab/arcasHLA
NeoPredPipe	Neoantigen prediction	MathOnco/NeoPredPipe

System Requirements

• Python 3.7+
• Java-enabled platform (Windows, Linux, macOS)
• 1–16 GB RAM

Background

Submitted as an MSc Bioinformatics thesis at Cardiff University. The pipeline integrates variant calling, HLA typing, and neoantigen prediction tools to produce a ranked list of candidate neoantigens from tumour RNA-seq data.