Dear friends,
The RNASeq data were aligned using tophat to get the the BAM files(which could be downloaded from SynapseClient). When I got the bam files, I used htseq-count to get the raw-counts for my subsequent DESeq analysis. However, I got all zeroes. Anyone here knows why? Any suggestions would be appreciated.
My terminal commands:
```
samtools sort -n -o 366_120502.sorted.bam 366_120502.bam
samtools view -@ 4 -bf 1 ./366_120502.sorted.bam -o ./366_120502.sorted.pe.bam
htseq-count -f bam --stranded=no --mode=intersection-nonempty -r name ./366_120502.sorted.pe.bam ~/mytools/cDNA/gencode14/gencode.v14.annotation.gtf.gz > ./htseq.366_2.txt
```
After running htseq-count:
33957780 SAM alignment pairs processed.
In htseq.366_2.txt:
$head htseq.366.txt
ENSG00000000003.10 0
ENSG00000000005.5 0
ENSG00000000419.8 0
ENSG00000000457.8 0
ENSG00000000460.11 0
ENSG00000000938.7 0
ENSG00000000971.11 0
ENSG00000001036.8 0
ENSG00000001084.6 0
ENSG00000001167.10 0
$head htseq.366_2.txt
ENSG00000000003.10 0
ENSG00000000005.5 0
ENSG00000000419.8 0
ENSG00000000457.8 0
ENSG00000000460.11 0
ENSG00000000938.7 0
ENSG00000000971.11 0
ENSG00000001036.8 0
ENSG00000001084.6 0
ENSG00000001167.10 0
Created by Yuan SHANG Raymond2018 I figured out the problem. It's all about the gft file.
As in the header of BAM file, there is no 'chr' before chromosome numbers. So, accordingly, we need to remove this part in the gtf file:
sed -e "s/^chr//g" gencode.v14.annotation.gtf > gencode.v14.annotation.no_chr.gtf
and use the no_chr.gtf file as the annotation files.
(Credits to akahles:
https://github.com/ratschlab/spladder/issues/29
)
Regards,