iPRISM User Guide"

In iPRISM: Intelligent Predicting Response to Cancer Immunotherapy Through Systematic Modeling

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

# Load the package
library(iPRISM)

Introduction

Welcome to the vignette for the PRISM package. This document provides an overview of the package's functionalities, including basic usage examples and detailed explanations of the main functions. The PRISM package includes the core function for the paper named PRISM: Predicting Response to cancer Immunotherapy through Systematic Modeling.

Example 1: Correlation Plot with Cell Types and Pathways

The cor_plot function generates a correlation plot between cell types and pathways, displaying correlation coefficients as a heatmap and significant correlations as scatter points.

# Read cell line and pathway information
data(data.path, package = "iPRISM")
data(data.cell, package = "iPRISM")

# Draw the plot
cor_plot(data1 = data.path, data2 = data.cell, sig.name1 = "path", sig.name2 = "cell")

Example 2: Enrichment Analysis using Multiplex Networks

The get_gsea_path function constructs a multiplex network, performs random walk restart, and calculates gene scores. It then transforms the scores and applies GSEA using the provided gene sets.

# Load example data
data(Seeds, package = "iPRISM")
data(ppi, package = "iPRISM")
data(path_list, package = "iPRISM")

# Shrink pathway list to the top 2 pathways
path_list <- path_list[1:5]

# Perform GSEA
result <- get_gsea_path(seed = Seeds, network = ppi, pathlist = path_list, gsea.nperm = 100)
print(result)

Example 3: Fit Logistic Regression Model

The get_logiModel function fits a logistic regression model as the paper highlighted, with an option for stepwise model selection.

# Load example data
data(data_sig, package = "iPRISM")

# Fit logistic regression model
b <- get_logiModel(data.sig = data_sig, pred.value = pred_value, step = TRUE)
summary(b)

iPRISM documentation built on Sept. 11, 2024, 7:14 p.m.