cluster.regions: Cluster samples using their abnormal regions

In jmonlong/PopSV: Population-based detection of structural variants from Read-Depth signal

Description

Cluster samples using their abnormal regions. The similarity between two samples is computed as the amount of base pairs called in both samples divided by the average of the total base pairs called in each sample. The distance is 1 - this similarity.

Usage

cluster.regions(cnv.df, cl.method = "complete", nb.cores = 3)

Arguments

cnv.df	a data.frame with the abnormal regions for each sample. Columns 'chr', 'start', 'end' and 'sample' are required.
cl.method	the linkage criterion for hierarchical clustering. Default is 'complete'.
nb.cores	the number of processors to use. Default is 1.

Value

a list

d	a distance matrix with the distance between each pair of sample.
hc	a 'hclust' object with the clustered samples.
cnv.geom	if the geometric approximation was used, the geometric representation of each samples as a matrix.

Author(s)

Jean Monlong

jmonlong/PopSV documentation built on Sept. 15, 2019, 9:29 p.m.