hey everyone,
I am trying to run the following code for C. elegans paired end data using the following code:


library("biomartr", lib.loc="~/R/x86_64-pc-linux-gnu-library/3.4")
library(BSgenome.Celegans.UCSC.ce11)
export(Celegans, "dataset/ce11.fasta", compress=FALSE)
getGFF(db = "ensembl", organism = "Caenorhabditis elegans", reference = TRUE, path = file.path("dataset"))
system("gunzip dataset/Caenorhabditis_elegans.WBcel235.91_ensembl.gff3.gz")
library(systemPipeR) 
args <- systemArgs(sysma="./param/tophat.param", mytargets="targetsPE.txt") 
system("bowtie2-build ./dataset/ce11.fasta ./dataset/ce11.fasta")
bampaths <- runCommandline(args=args)
read_statsDF <- alignStats(args)
write.table(read_statsDF, "results/alignStats.xls", row.names=FALSE, quote=FALSE, sep="\t")
txdb <- makeTxDbFromGFF(file="dataset/Caenorhabditis_elegans.WBcel235.91_ensembl.gff3.gz", format="gff")
saveDb(txdb, file="dataset/database.sqlite")
txdb <- loadDb("./dataset/database.sqlite")

eByg <- exonsBy(txdb, by=c("tx"))
countDFeByg <- assays(counteByg)$counts
countDFeByg <- as.data.frame(countDFeByg)
countDFeByg[is.na(countDFeByg)] <- 0
write.table(countDFeByg, "results/countDFeByg.xls", col.names=NA, quote=FALSE, sep="\t")
countDF <- read.table("results/countDFeByg.xls")
cmp <- readComp(args, format="matrix", delim="-")
edgeDF <- run_edgeR(countDF=countDF, targets=targetsin(args), cmp=cmp[[1]], independent=FALSE, mdsplot="")
write.table(edgeDF, "results/edgeRcomp.xls", quote=FALSE, sep="\t", col.names = NA)
edgeDF <- read.table("./results/edgeRcomp.xls")
DEG_list <- filterDEGs(degDF=edgeDF, filter, plot = FALSE)

while running the following code i am facing the error as below
     Error in calcNormFactors.default(object = object$counts, lib.size = object$samples$lib.size,  : 
     NA counts not permitted
     Called from: calcNormFactors.default(object = object$counts, lib.size = object$samples$lib.size,
     method = method, refColumn = refColumn, logratioTrim = logratioTrim, sumTrim = sumTrim, 
     doWeighting = doWeighting, Acutoff = Acutoff, p = p)

I have already tried doing 

countDFeByg[is.na(countDFeByg)] <- 0

which is said to remove the NA.

It would be very kind if someone may help me to solve the error or point out where I am going wrong

Thanks in advance.

Claris

Obviously, whatever matrix is being passed to calcNormFactors has NA counts. Removing NAs in countDFeByg won't help if the NAs are being introduced afterwards.

• Are the NAs present in countDF? What does anyis.na(countDF)) say?
• Do any of the libraries have zero counts? This will result in NAs if run_edgeR() computes CPMs.

Otherwise, it's something happening inside run_edgeR() (I assume this is where the error is being raised), and you'll have to do debug(run_edgeR) to see how the NAs are being introduced.