CALCULATE DESCRIPTIVE AND TEST STATISTICS

Description

Calculate the (difference in) median beta and M values for the rows specified in probe_rows (descriptive statistics). probe_rows must specify a continuous range of adjacent probes. If a regions shows an high difference (>dM_min) between any pair of samples, the differences are formally tested (Mann Whitney U if n=2, Kruskal Wallis n>2). If n>2 and the omnibus test is significant , pairwise testing is performed to determine which differ significantly.

Usage

1
calc_stats(probe_rows,probes,m,b,n,dMth,do.format=FALSE)

Arguments

probe_rows

Vector of row numbers in regions$valid.m, regions$valid.beta or regions$perprobe identifying adjacent probes. See output regionfinder

probes

see output of regionfinder (regions$perprobe)

m

see output of regionfinder (regions$valid.m)

b

see output of regionfinder (regions$valid.beta)

n

total number of samples

dMth

see min_dM parameter of DMRforPairs

do.format

see testregion

Details

The p-values reported by this function are not yet adjusted for multiple testing. This function is rarely called by the user directly.

Value

Data frame or vector with statistics and test results concerning the probes specified in probe_rows (please see description).

Author(s)

Martin Rijlaarsdam, m.a.rijlaarsdam@gmail.com

See Also

DMRforPairs, regionfinder, testregion.

Examples

1
2
3
#calc_stats() is an integrated part of the DMRforPairs() wrapper and is 
#not usually called by the user directly. Please see DMRforPairs() for 
#an example.