psSubpathway (version 0.1.3)

DCSA: Dynamic Changed Subpathway Analysis (DCSA)

Description

This function will perform the Dynamic Changed Subpathway Analysis (DCSA) method to estimate the dynamic changed subpathways associated with the sample phenotypes (like the developmental stage of cancer).

Usage

DCSA(
  expr,
  input.cls = "",
  subpathwaylist = "Symbol",
  kcdf = "Gaussian",
  method = "gsva",
  min.sz = 5,
  max.sz = 1000,
  nperm = 100,
  fdr.th = 1,
  mx.diff = TRUE,
  parallel.sz = 0
)

Value

A list containing the results of DCSA and subpathway activity profile.

Arguments

expr

Matrix of gene expression values (rows are genes, columns are samples).

input.cls

Input sample phenotype class vector file in CLS format.

subpathwaylist

Character string denoting the gene label of the subpathway list is `Entrezid` (default) or `Symbol`. Users can also enter their own subpathway list data. This list should be consistent with the gene label in the input gene expression profile.

kcdf

Character string denoting the kernel to use during the non-parametric estimation of the cumulative distribution function of expression levels across samples when method="gsva". By default, `kcdf="Gaussian"` which is suitable when input expression values are continuous, such as microarray fluorescent units in logarithmic scale, RNA-seq log-CPMs, log-RPKMs or log-TPMs. When input expression values are integer counts, such as those derived from RNA-seq experiments, then this argument should be set to `kcdf="Poisson"`.

method

Method to employ in the estimation of subpathway enrichment scores per sample. By default, this is set to `gsva` (Hänzelmann et al, 2013) and other options are `ssgsea` (Barbie et al, 2009).

min.sz

Minimum size of the resulting subpathway.

max.sz

Maximum size of the resulting subpathway.

nperm

Number of random permutations (default: 100).In practice, the users can set their own values as needed, and more than 1000 times may be fine in general.

fdr.th

Cutoff value for FDR. The only subpathway with lower FDR.th are listed (default: 1).

mx.diff

Offers two approaches to calculate the sample enrichment score (SES) from the KS random walk statistic. `mx.diff=FALSE`: SES is calculated as the maximum distance of the random walk from 0. `mx.diff=TRUE` (default): SES is calculated as the magnitude difference between the largest positive and negative random walk deviations.

parallel.sz

Number of processors to use when doing the calculations in parallel. If this argument is left with its default value (parallel.sz=0) then it will use all available core processors unless we set this argument with a smaller number.

Author

Xudong Han, Junwei Han, Qingfei Kong

Details

DCSA

This function calculates the subpathway activity profile based on the gene expression profile and subpathway list by 'gsva' or 'ssgssea'. Next, we used the information-theoretic measure of statistical dependence, mutual information (MI), to estimate the dynamically changed subpathways associated with the sample phenotypes. Finally we used the perturbation analysis of the gene label rearrangement to estimating the statistical significance.

Examples

# load depend package.
require(GSVA)
require(parallel)
require(mpmi)
# get ACC disease stage gene expression profiling.
# ACCgenematrix<-get("DCgenematrix")
# get path of the sample disease stage phenotype files.
# Stagelabels<-system.file("extdata", "DClabels.cls", package = "psSubpathway")
# perform the DCSA method.
# DCSA(ACCgenematrix,input.cls=Stagelabels,nperm=50,fdr.th=0.01,parallel.sz=2)
# get the result of the SubSEA function
# DCSAresult<-get("DCspwresult")
# str(DCSAresult)
# head(DCSAresult$DCSA)

# Simulated gene matrix.
genematrix <- matrix(rnorm(500*40), nrow=500, dimnames=list(1:500, 1:40))
# Construct subpathway list data.
subpathwaylist <- as.list(sample(2:100, size=20, replace=TRUE))
subpathwaylist <- lapply(subpathwaylist, function(n) sample(1:500, size=n, replace=FALSE))
names(subpathwaylist)<-c(paste(rep("spw",20),c(1:20)))
# Construct sample labels data.
stagelabel<-list(phen=c("stage1","stage2","stage3","stage4"),
                   class.labes=c(rep("stage1",10),rep("stage2",10),
                   rep("stage3",10),rep("stage4",10)))
DCSAcs<-DCSA(genematrix,stagelabel,subpathwaylist,nperm=0,parallel.sz=1)