ct.parseGeneSymbol: Parse 'geneSymbol' values to construct an alternative...

In OscarBrock/gCrisprTools: Suite of Functions for Pooled Crispr Screen QC and Analysis

Parse 'geneSymbol' values to construct an alternative annotation list

Description

This is an accessory function to 'ct.expandAnnotation()' function, which enables users to expand annotation objects to accomodate reagent libraries where reagents are expected to impact a set of known targets. See documentation for that function for additional details.

Often, libraries that contain multiply-targeting reagents are annotated using a structured format that can be decomposed by regex matching. This function takes in an annotation object containing an 'ID' column indicating the reagent ID and a 'geneSymbol' column containing the target mappings, and parses the target mappings according to a known annotation format. Currently supported formats are 'cellecta' (e.g., 'TARGET_P1P2P3' indicating multiple promoters associated with a known target), and 'underscore', where different targets are concatenated using the underscore separator (e.g., 'TARGET1_TARGET2_TARGET3').

Returns an 'alt.annotation'-type list of character vectors encoding the target mappings for each reagent.

Usage

ct.parseGeneSymbol(ann, format = c("cellecta", "underscore"))

Arguments

ann	A data.frame containing reagent-level information encoded as rows. The 'ID' column should correspond to the individual reagent identifiers, and the 'geneSymbol' column should contain target annotation strings to be parsed (both are coerced to strings). Does not, strictly speaking, need to be a proper annotation object, but one of those will work.
format	Format of the geneSymbol column strings.

Value

A named 'alt.annotation'-type list of character vectors encoding the target mappings for each reagent

Author(s)

Russell Bainer

Examples

fakeann <- data.frame('ID' = LETTERS[1:4], 'geneSymbol' = c('T1_P1', 'T1_P1P2', 'T1_P2P1', 'T1_P2'))
ct.parseGeneSymbol(fakeann, 'cellecta')
ct.parseGeneSymbol(fakeann, 'underscore')

OscarBrock/gCrisprTools documentation built on Oct. 25, 2022, 7:29 a.m.