In hmumme/survivalHelp: What the Package Does (One Line, Title Case)

knitr::opts_chunk$set(echo = TRUE)
#devtools::install("hmumme/survivalHelp") # uncomment if you have not installed survivalHelp
library(survivalHelp)
load("~/GitHub/survivalHelp/demo/data/mpalData.RData") # load demo data

The file mpalData.RData contains the necessary objects to run survival analysis on the MPAL samples from the TARGET-ALL-P3 (ALAL) datasets. It includes the bulkRNA-seq expression matrices for both the B/Myeloid and T/Myeloid pediatric MPAL samples, the ensmbl id matrices, the ensmbl to gene symbol reference data frame, and the survival data for both subsets.

ls()

Single Gene

First, we will show how to obtain survival plots for a single gene's expression in B/Myeloid MPAL samples. You can use either median or cutP as the cut method here.

gene = "TP53"
out = mCut(gene, bMat, osB, cut = "cutP")
out

The mCut function returns the gene expression cutoff to use to split samples into High and Low groups based on cutP method and the gene expression values for each sample.

osB["group"] = cutGroups(osB, out)
osB

The samples have been split into High and Low categories in the osB group column.

plotSurv(osB, name = "TP53")

We can also calculate the survival statistics.

survStats(osB)

We calculated the logRank p-value (survdiffP) for the survival plot, the Cox Hazard Ratio (coxHR), and the Cox HR p-value (coxP). From these statistics, we can see that the expression of TP53 does not have a significant effect on B/Myeloid MPAL pediatric survival.

Gene Set

Second, we will show how to obtain survival plots for a combined gene set's expression in T/Myeloid MPAL samples. You can use either median or cutP as the cut method here. Instead of individual expression values, we take the GSEA scores for a gene set and use this when splitting samples into High/Low groups.

gs = c("TP53","ICAM1","HPGD")
out = mSet(gs = gs, ENS = tENS, os = osT, ref = ref, cut = "cutP")
out

The mSet function returns the GSEA score cutoff to use to split samples into High and Low groups based on cutP method and the GSEA values for each sample.

osT["group"] = cutGroups(osT, out)
osT

The samples have been split into High and Low categories in the osT group column.

plotSurv(osT, name = "TP53, ICAM1, and HPGD")

We can calculate the survival statistics for the gene set

survStats(osT)

We calculated the logRank p-value (survdiffP) for the survival plot, the Cox Hazard Ratio (coxHR), and the Cox HR p-value (coxP). From these statistics, we can see that the gene set of TP53, ICAM1, and HPGD does not have a significant effect on T/Myeloid MPAL pediatric survival.

hmumme/survivalHelp documentation built on June 6, 2022, 9:15 p.m.