clinical.ranksum: Compute associations between gene set/module and patient...

In vdumeaux/mixtR: Matched Interaction across Tissues (MIxT) data analysis

Description

Using ranksums to capture gene set/module expression, we can ask how gene sets/modules in each tissue are differentially expressed according to patient<e2><80><99>s clinicopathological variables. The type of the clinicopathological attribute (categorical or continuous) determines the underlying statistical test. Pearson correlation (Student asymptotic p-value) is used to test association between a given module and continuous patient attributes (eg. age). Analysis of Variance (ANOVA) is used to test association between a given module and categorical patient attributes (eg. ER status).

Usage

clinical.ranksum(mixt.dat, mixt.ranksum, tissue, cohort.name = "all",
  corType = "p", nRuns = 10000, randomSeed = 12345, mc.cores = 2,
  verbose = 2)

Arguments

mixt.dat	Data object from matched tissues.
mixt.ranksum	output of sig.ranksum()
tissue	character string that provides the name of tissue we want to look at
cohort.name	character string that provides the name of the patient cohort to analyze, default is set to 'all'
corType	a character string indicating which correlation coefficient is to be computed. Default 'p' for pearson
nRuns	number of permutations
randomSeed	seed number for random number generation. Default set as '1234'
mc.cores	number of cores
verbose	numerical. default > 0 show informational text on progress

Value

matrix of p-values determining significance of association between clinical variables (in row) and gene sets/modules (in column)

vdumeaux/mixtR documentation built on May 3, 2019, 4:58 p.m.