HTSeq-counts can not map any reads to gtf file

Question

[nazaninhoseinkhan]

Dear all,

I need to get the raw read counts of a *pseudomonas aeruginosa UCBPP-PA14 * RNASeq data set.

After mapping with "Bowtie for Illumina", I run HTSeq-counts using 2 version of pseudomonas aeruginosa UCBPP-PA14 gft files obtained from ensembl and pseudomonas genome databases, however htseq-counts resulted in to empty file telling that no reads could be mapped.

Similarly I tried featureCounts , but again it could not map any read.

Can you advice me to solve this problem?

Regards

Nazanin

[ra]

Have you had a look at your .bam file in IGV? This will allow you to visualise the alignments from Bowtie and indicate if the problem lies at your mapping stage.

Answer 1

[Friederike]

A typical problem is that the chromosome names don't match between the reference sequence and the GTF file.

Try running:

samtools idxstats my.bam

The output should look something like this:

chr1    195471971   1866052 0
chr2    182113224   1667812 0
chr3    160039680   1339058 0
chr4    156508116   1318268 0
chr5    151834684   1383842 0

where the columns correspond to the names of the chromosome, their length, the number of mapped reads and the number of unmapped reads.

Make sure that a) you have mapped reads and b) that the chromosome names are exactly the same as those in the first column of the GTF file that you use with featureCounts/HTSeqCounts.

[nazaninhoseinkhan]

Hi,

Thank u for your reply

The chromosome names are the same in both reference and gtf file: "chromosome".

The out put of running IDxstats on my bam file was:

1 2 3 4

chromosome 6537648 1021248 0

• 0 0 2584368

I am pretty sure that I have mapped the reads since I had previously used the result of mapping in cuffdiff and I got the results.

Since I have to remove the batch effects between my samples I have to run deseq2 and I needs the raw read counts