variantFilter: Class '"variantFilter"'

In dbottomly/HitWalker: Network-based variant prioriziation using functional assay scores

Description

This class provides a way of working with text-based variant filters such as those provided in the Genome Analysis Toolkit. Most importantly, users can specify labels for the filters and the type of variant to which they should be applied.

Objects from the Class

Objects should be created from a call to the function makeVariantFilter. See the example section.

Slots

filter.dta:

Object of class "data.frame" containing three columns: 'category', 'label' and 'type', all of character mode. 'Category' refers to how the end user should view the filter value (e.g. PASS or FAIL). The data.frame should be sorted based on decreasing category value. 'Label' refers to the actual filters, note multiple filters applied to a variant will be parsed using filter.sep. All combinations of the filters will be considered internally so only one combination needs to be specified. 'Type' refers to the type of variant that the filter(s) supplied in label should be applied to (e.g. INDEL or SNV). Again can refer to multiple variants using filter.sep.

filter.sep:

Object of class "character" delimiting multiple labels or types.

filter.collapse.col:

Object of class "character" containing the name of the column from the retrieved variant data that the values in label would apply to.

filter.ignore.col:

Object of class "character" indicating redudant columns with filter.collapse.col. This could be an auto-incremeted integer value produced by a DBMS.

filter.best.category:

Object of class "character" containing the 'category' that should be considered the best quality.

Methods

filterBestCategory

signature(obj = "variantFilter"): A method for retrieving the best category

filterCategory

signature(obj = "variantFilter"): A method for retrieving a character vector of categories named by label

filterCollapseCol

signature(obj = "variantFilter"): A method for retrieving the column of the variant data.frame that contains the variant filters

filterDta

signature(obj = "variantFilter"): A method to retrieve the filter data.frame

filterIgnoreCol

signature(obj = "variantFilter"): A method to retrieve the columns of the variant data.frame that should be ignored

filterLabels

signature(obj = "variantFilter"): A method to retrieve a character vector containing filter types named by filter labels

filterSep

signature(obj = "variantFilter"): A method for retrieving the character used to seperate the label and type values

Author(s)

Daniel Bottomly

References

Genome Analysis Toolkit Website: http://www.broadinstitute.org/gatk/

See Also

priorDbParams, get.sample.variants, filter.variant.annots

Examples

category <- c("OK", "OK", "WARN", "WARN", "WARN", "WARN")
label <- c("PASS", "StrandBiasFilter", "GATKStandardFilter", "HARD_TO_VALIDATE", "HARD_TO_VALIDATE;GATKStandardFilter", "GATKStandardFilter;StrandBiasFilter")
type <- c("SNV;INDEL", "SNV;INDEL", "SNV;INDEL", "SNV;INDEL", "SNV;INDEL", "INDEL")
    
filter.dta <- data.frame(category, label, type, stringsAsFactors=FALSE)
var.filter <- makeVariantFilter(filter.dta, filter.sep=";", filter.collapse.col="filter", filter.ignore.col="variant_qual_id", filter.best.category="OK")

dbottomly/HitWalker documentation built on May 15, 2019, 1:22 a.m.