New Rscript with comments and better logo length control

#!/usr/bin/env Rscript

#!/usr/bin/env Rscript

library(seqLogo)

library(Biostrings)

# Get command line arguments

args = commandArgs(trailingOnly=TRUE)

args = commandArgs(trailingOnly=TRUE)

# Set variables based on command arguments

filename <- args[1]

filename <- args[1]

logolen = 0

if(args[2] > 0) {

  logolen = args[2]

}

filebase <- strsplit(filename,split="\\.")[[1]][1]

filebase <- strsplit(filename,split="\\.")[[1]][1]

library(seqLogo)

# Read in input sequences from input filename (=args[1])

library(Biostrings)

dna <- readDNAStringSet(filename)

dna <- readDNAStringSet(filename)

numseq = length(dna)

# Align input sequences

cm <-consensusMatrix(dna)[1:4,]

cm <-consensusMatrix(dna)[1:4,]

# Count number of aligned bases (without indels) at each position

sumcm <- apply(cm,2,sum)

sumcm <- apply(cm,2,sum)

cm[1,] <- cm[1,]/sumcm

cm[1,] <- cm[1,]/sumcm

cm[2,] <- cm[2,]/sumcm

cm[2,] <- cm[2,]/sumcm

cm[3,] <- cm[3,]/sumcm

cm[3,] <- cm[3,]/sumcm

cm[4,] <- cm[4,]/sumcm

cm[4,] <- cm[4,]/sumcm

if(length(cm[1,]) > 16) {

  cm <- cm[,1:16]

# If consensus matrix too wide, shorten to logolen (=args[2])

if(length(cm[1,]) > logolen) {

  cm <- cm[,(which(sumcm>numseq-1)[1]+1):(which(sumcm>numseq-1)[1]+logolen)]

}

}

pwmcm <- makePWM(cm)

# Create sequence logo and save it as PNG file

pwmcm <- makePWM(cm)

png(file=paste0(filebase,".png"), width=480, height=240)

png(file=paste0(filebase,".png"), width=480, height=240)

seqLogo(pwmcm)

seqLogo(pwmcm)

dev.off()

dev.off()