make.pwm: Construct position-weight matrix for subsets of reads from...

In mw55309/viRome_legacy: Analysis and visualisation of short-read sequencing data from virus-infection studies

Description

This function takes a data frame as input, which should be identical to the output of read.bam and clip.bam. It filters according to the length of the mapped reads, and then countsthe occurrence of each base along the length of the read(s). The data are then optionally scaled.

Usage

make.pwm(vdf = NULL, minlen = 1, maxlen = 37, scaled = TRUE, strand = "pos", revcom = FALSE, ttou = FALSE)

Arguments

vdf	Data frame, should be the output of the output of clip.bam
minlen	The minimum length of mapped read to include
maxlen	The maximum length of mapped read to include
scaled	Whether or not to scale each column of base counts to the total number of bases in that column. Default: TRUE
strand	The strand to calculate the PWM on: either "pos" or "neg"
revcom	Whether or not to reverse-complement the sequence. We recommend you leave this as default. The default should work except those times when, for example, a negative strand virus has been published as a positive strand, and you have aligned your data to the positive strand.
ttou	Whether or not to convert T to U (Uracil). Currently not recommended. In all liklihood, you measured cDNA anyway, not RNA, and therefore you should report a T :)

Value

A matrix with the c("A","G","C","T") as rows, the position as columns and the (scaled) counts as values

Author(s)

Mick Watson

Examples

## Not run: infile <- system.file("examples/SRR389184_vs_SINV_sorted.bam", package="viRome")
 ## Not run: bam <- read.bam(bamfile=infile, chr="SINV", minlen=1, maxlen=11703, removeN=TRUE)
 ## Not run: bamc <- clip.bam(bam)
 ## Not run: make.pwm(bamc, minlen=25, maxlen=37)

mw55309/viRome_legacy documentation built on Dec. 21, 2021, 11:05 p.m.