Calculating abundance for each bin

[Camilla Nesbø]

Is there a way to get median depth for each bin? To use as abundance? (i.e. is there already a script or function for this) This would be a great addition to the output.

[Don Kang]

Thanks for the suggestion. Meanwhile, you can calculate it with depth file as shown below (R code). Make it sure to have correct depth file and bin directory.

library(foreach)
library(matrixStats)

depth <- read.table('depth.txt', as.is=T, header=T)
depth <- depth[,-which(grepl('var$',colnames(depth)))][,-c(2,3)]
bins <- list.files('bins_dir', pattern='.fa', full.name=T)
mems <- foreach(f = bins) %do% {
    sub('>','',system(sprintf("grep '^>' %s", f), intern=T))
}
names(mems) <- sub('.fa','',basename(bins)) 

bins.cv <- do.call(rbind, foreach(m=mems) %do% {
    colMedians(as.matrix(depth[match(m, depth$contigName),-1]))
})
dimnames(bins.cv) <- list(names(mems), colnames(depth)[-1])
write.table(bins.cv, file='MedianCoverage.txt', row.names=T, col.names=T, quote=F, sep='\t')

[Camilla Nesbø]

thanks!