Hello,
After counting genomic features with featureCounts
featureCounts -F GTF -t exon -g gene_id -f -O --minOverlap 2 -M --fraction -J -a annotation.gtf -o out.txt 1_sorted.bam 2_sorted.bam 3_sorted.bam 4_sorted.bam
I decided to use the output file as an input for edgeR to create the DGElist.
First I used the read.delim
function in R to read the text file and create a table. The command I used is shown below.
counts <- read.delim("~/Desktop/thesis_bioinformatics/chick_fl_hl_comparison/counts_fcounts.txt", sep="\t", header = TRUE, check.names = FALSE, stringsAsFactors = FALSE)
After getting the table, I tried using it to create the DGElist by using the commands below.
group<-c(1,1,2,2)
y <- DGEList(counts=counts, group=group)
This outputs an error message that says Error: Negative counts not allowed
. I followed someones advice from this post Question: Error when running edgeR, where they suggest using is.na(counts) || counts < 0
to check if NA or negative counts are present. I checked all the columns of my counts
file and the only column where the output was true was the Strand column, which describes the directionality. I removed this column from the table and tried running DGElist again with a counts file that did not have the Strands column. The same error message appeared.
I'm pretty confused as to why I get this error message when none of my counts are negative. Any help would be appreciated.
Cheers, Fjodor
The start of my counts
table looks something like this:
Geneid Chr Start End Strand Length 1 2 3 4
1 ENSGALG00000054818 1 9774 10061 - 288 # # # #
Did you manipulate
out.txt
prior to reading it with theread.delim
command? With this command you should not be able to read afeatureCounts
output. It should complain about the header line (starting with# Program....
) not matching the number of columns. Please showhead -n 10 counts_fcounts.txt
Yes, I did remove the header and gave the counts columns easier names.