sv.summary: Abnormal regions summary

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

Description

Summary of regions with abnormal coverage as detected by 'call.abnomal.cov'. It's a static version of 'sv.summary.interactive'. It produces graphs to to get an idea of the quality of the calls. For normal samples we would expect: the amount of called genome in each sample to be somewhat similar; the copy number estimates to locate near integer values; No systematic calls (i.e. called in all samples). The output graphs represent the amount of affected genome across samples, the copy number estimates distribution, the frequency distribution.

Usage

sv.summary(res.df, out.pdf = NULL, print = TRUE, FDR.th = 0.05,
  min.cn2Dev = 0, min.cov = 0, max.sing.kb = Inf)

Arguments

res.df	the data.frame with the results.
out.pdf	the name for the output PDF file to create. If NULL (default), the graphs are either displayed directly or returned as a list.
print	Should the graphs be displayed ? If 'out.pdf' though, this parameter is not used as no graph will be displayed.
FDR.th	the FDR threshold to use. Default is 0.05.
min.cn2Dev	the minimum deviation from CN2. Default is 0. Using '0.5' would remove CNVs with CN estimates in [1.5,2.5].
min.cov	the minimum coverage in the reference samples. Default is 0.
max.sing.kb	the maximum amount of single-bin calls. Default is Inf. Use and tweak when batches or outliers are visible in the calls.

Details

A list is returned. Its first element is a list of ggplot object with the different graphs. The second is a data.frame with summary information used to create the graphs.

Value

A list with :

graphs.l	a list of ggplot graphs
cnv.df	a data.frame with the selected CNVs
info.df	a data.frame with different summary statistics

Author(s)

Jean Monlong

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