tigeR

tigeR is an R package designed for exploring biomarkers and constructing predictive models for immunotherapy response via built-in or custom immunotherapy gene expression data.

1. Introduction

• Built-in datasets: 1060 samples with immunotherapy clinical information from 11 melanoma datasets, 3 lung cancer datasets, 2 kidney cancer datasets, 1 gastric cancer dataset, 1 low-grade glioma dataset, 1 glioblastoma dataset and 1 head and neck squamous cell cancer dataset (all organized into R language ‘SummarizedExperiment’ objects).

• 23 immunotherapy response-related biomarkers from literature, multiple methods for analysis and visualization.

• 10 open source tumor microenvironment deconvolution methods including CIBERSORT, TIMER, ESTIMATE, IPS, xCell, EPIC, ConsensusTME, ABIS, quanTIseq, and MCPCounter. Several downstream method for analysis and visualization.

• 7 machine learning method for multi-modal prediction model construction and testing.

<p align="center"> <img src="https://raw.githubusercontent.com/Chengxugorilla/tigeR/gh-pages/figs/Figure 1.svg" width="80%"> </p> <p align="center"><b>Overall design of tigeR</b></p>

2. Installation

packages <- c("BiocManager", "devtools", "ggplot2", "pROC", "RobustRankAggreg")
for (package in packages) {
  if (!require(package, character.only = TRUE)) {
    install.packages(package)
  }
}
devtools::install_github("YuLab-SMU/tigeR")

3. Quick Start

The workflow of tigeR is below, see more details in tigeR documentation.

<p align="center"> <img src="https://raw.githubusercontent.com/Chengxugorilla/tigeR/gh-pages/figs/Figure 2.svg" width="80%"> </p> <p align="center"><b>Overall design of tigeR</b></p>

4. TIGER web server

TIGER Web Server