Rapid Identification of Genomic Alterations in Tumors affecting lymphocyte Infiltration (RIGATonI)

Abstract

Recently identified tumor genomic alterations are associated with altered tumor immune microenvironments and therapeutic outcomes. These studies beg the question: are there other genomic variations altering the immune microenvironment in tumors that can provide insight into mechanisms of immune evasion? Current approaches to estimate immunity in bulk RNAseq from tumors include gene set enrichment and cellular deconvolution. In contrast, our new software package RIGATonI utilizes a gene agnostic approach to classify immune cell infiltration in tumors. Using pathologist reviewed histology slides and paired bulk RNA sequencing expression data, we trained a machine learning algorithm to detect inflamed and non-inflamed immune microenvironments. RIGATonI enables identifcation of candidate genomic alterations associated with immune infilration in subsets of cancer through evaluation of various features including gene expression, genomic alterations, pathologist-classified tumors, clinical outcomes, and protein validation. In addition to immune infiltrate classification, RIGATonI leverages a novel algorithm using known protein-protein interactions for prediction of gain-of-function and loss-of-function for genomic alterations to improve filtering of genomic alteration candidats. Applying these two novel methods, we analyzed all genomic alterations present in The Cancer Genome Altas and visualized promising results with R shiny for public use. Thus, we present our R tool and online database for Rapid Identification of Genomic Alterations in Tumors affecting lymphocyte Infiltration (RIGATonI).

Competing Interest Statement

The authors have declared no competing interest.