cnvs_inheritance: Compute CNVs inheritance

In SinomeM/CNVgears: A Framework to Combine, Analize and Interpret CNVs Calling Results

Description

cnvs_inheritance compute CNVs inheritance pattern and search good de novo CNVs candidates

Usage

cnvs_inheritance(
  sample_list,
  markers,
  results,
  raw_path,
  mmethod = 1,
  alfa = 0.05,
  min_NP = 10,
  adjust_pval = TRUE,
  reciprocal_overlap = 0.3
)

Arguments

sample_list	a data.table, the output of read_metadt.
markers	a data.table, the output of read_NGS_intervals or read_finalreport_snps, depending on the initial data type.
results	a CNVresults object, the output of read_results or inter_res_merge.
raw_path	character, the path where raw data processed by read_NGS_raw or read_finalreport_raw.
mmethod	fine-screening method.
alfa	p value alfa value, default in 0.05.
min_NP	minimum number of points to try fine screening a de novo candidate
adjust_pval	logical compute adjusted p-value?
reciprocal_overlap	minimum reciprocal overlap for inherited CNVs detection

Details

Given a trio (mother, father, offspring) this function computes inheritance patterns of the offspring's CNVs. This is done both by comparing the actual calls in the trio (CNVs-level) and the raw data (markers-level). For this reason it is suggested to use it separately on each method and then combine the results. As an example if the focus are de novo CNVs, select only those and then combine the outputs using inter_res_merge. In this way it also possible to further increase the confidence in the results, e.g. by filtering out all de novo candidates called by a single method.

Internally the function is structured in several steps. [...]

Value

a CNVresults object with some additional columns: inheritance and related p-values.

SinomeM/CNVgears documentation built on Nov. 21, 2021, 5:34 a.m.