R: Trim ends of reads based on nucleotides or qualities

trimTails {ShortRead}

R Documentation

Trim ends of reads based on nucleotides or qualities

Description

These generic functions remove leading or trailing nucleotides or qualities. trimTails and trimTailw remove low-quality reads from the right end using a sliding window (trimTailw) or a tally of (successive) nucleotides falling at or below a quality threshold (trimTails). trimEnds takes an alphabet of characters to remove from either left or right end.

Usage

trimTailw(object, k, a, halfwidth, ..., ranges=FALSE)

## S4 method for signature 'BStringSet'
trimTailw(object, k, a, halfwidth, ..., alphabet, ranges=FALSE)
## S4 method for signature 'XStringQuality'
trimTailw(object, k, a, halfwidth, ..., ranges=FALSE)

trimTails(object, k, a, successive=FALSE, ..., ranges=FALSE)
## S4 method for signature 'BStringSet'
trimTails(object, k, a, successive=FALSE, ...,
    alphabet, ranges=FALSE)
## S4 method for signature 'XStringQuality'
trimTails(object, k, a, successive=FALSE, ..., ranges=FALSE)

trimEnds(object, a, left=TRUE, right=TRUE, relation=c("<=", "=="),
    ..., ranges=FALSE)
## S4 method for signature 'XStringSet'
trimEnds(object, a, left=TRUE, right=TRUE, relation=c("<=", "=="),
    ..., ranges=FALSE)
## S4 method for signature 'XStringQuality'
trimEnds(object, a, left=TRUE, right=TRUE, relation=c("<=", "=="),
    ..., ranges=FALSE)
## S4 method for signature 'FastqQuality'
trimEnds(object, a, left=TRUE, right=TRUE, relation=c("<=", "=="),
    ..., ranges=FALSE)
## S4 method for signature 'ShortRead'
trimEnds(object, a, left=TRUE, right=TRUE, relation=c("<=", "=="),
    ..., ranges=FALSE)
## S4 method for signature 'ShortReadQ'
trimEnds(object, a, left=TRUE, right=TRUE, relation=c("<=", "=="),
    ..., ranges=FALSE)

Arguments

`object`	An object for which a methods exist (e.g., `ShortReadQ` and derived classes); see below to discover these methods.
`k`	`integer(1)` describing the number of failing letters required to trigger trimming.
`a`	For `trimTails` and `trimTailw`, a `character(1)` with `nchar(a) == 1L` giving the letter at or below which a nucleotide is marked as failing. For `trimEnds` a `character()` with all `nchar() == 1L` giving the letter at or below which a nucleotide or quality scores marked for removal.
`halfwidth`	The half width (cycles before or after the current; e.g., a half-width of 5 would span 5 + 1 + 5 cycles) in which qualities are assessed.
`successive`	`logical(1)` indicating whether failures can occur anywhere in the sequence, or must be successive. If `successive=FALSE`, then the k'th failed letter and subsequent are removed. If `successive=TRUE`, the first succession of k failed and subsequent letters are removed.
`left, right`	`logical(1)` indicating whether trimming is from the left or right ends.
`relation`	`character(1)` selected from the argument values, i.e., “<=” or “==” indicating whether all letters at or below the `alphabet(object)` are to be removed, or only exact matches.
`...`	Additional arguments, perhaps used by methods.
`alphabet`	`character()` (ordered low to high) letters on which quality scale is measured. Usually supplied internally (user does not need to specify). If missing, then set to ASCII characters 0-127.
`ranges`	`logical(1)` indicating whether the trimmed object, or only the ranges satisfying the trimming condition, be returned.

Details

trimTailw starts at the left-most nucleotide, tabulating the number of cycles in a window of 2 * halfwidth + 1 surrounding the current nucleotide with quality scores that fall at or below a. The read is trimmed at the first nucleotide for which this number >= k. The quality of the first or last nucleotide is used to represent portions of the window that extend beyond the sequence.

trimTails starts at the left-most nucleotide and accumulates cycles for which the quality score is at or below a. The read is trimmed at the first location where this number >= k. With successive=TRUE, failing qualities must occur in strict succession.

trimEnds examines the left, right, or both ends of object, marking for removal letters that correspond to a and relation. The trimEnds,ShortReadQ-method trims based on quality.

Value

An instance of class(object) trimmed to contain only those nucleotides satisfying the trim criterion or, if ranges=TRUE an IRanges instance defining the ranges that would trim object.

Author(s)

Martin Morgan <mtmorgan@fhcrc.org>

Examples

showMethods(trimTails)

sp <- SolexaPath(system.file('extdata', package='ShortRead'))
rfq <- readFastq(analysisPath(sp), pattern="s_1_sequence.txt")

## remove leading / trailing quality scores <= 'I'
trimEnds(rfq, "I")
## remove leading / trailing 'N's
rng <- trimEnds(sread(rfq), "N", relation="==", ranges=TRUE)
narrow(rfq, start(rng), end(rng))
## remove leading / trailing 'G's or 'C's
trimEnds(rfq, c("G", "C"), relation="==")

[Package ShortRead version 1.12.4 Index]