strange htseq-count result

Question

[lilingjoyo]

hello all,

I'm using STAR to 2-pass alignment and htseq-count to output gene count table. But my htseq-count result table is a bit strange, like follow:

__no_feature    5802774

__ambiguous 4365152

__too_low_aQual 0

__not_aligned   0

__alignment_not_unique  0

all alignments are in no feature or ambiguous couters. Why does this happen. My code is:

htseq-count -s no Aligned.out.sorted.sam gencode.v28.annotation.gtf > samp.htseq-count.tab

Any one has got same problem?

[WouterDeCoster]

I added (code) markup to your post for increased readability. You can do this by selecting the text and clicking the 101010 button. When you compose or edit a post that button is in your toolbar, see image below:

[WouterDeCoster]

Does the gtf used for htseq-count contain the same chromosome identifiers as your alignment in sam format?

[lilingjoyo]

I thinks so. In my chrName.txt one of my star index files, the chromosome id is in the form of chr#. In my gencode gtf file, it's same.

[grant.hovhannisyan]

To save your time and shorten the pipeline, you can use STAR's --quantMode option (have a look in manual). It will generate exactly the same read-counts as htseq-count (note that probably this would not solve the issue you posted here).

Answer 1

[Matteo Schiavinato]

-s no

This is (imho) the problem. Perhaps your library is stranded. I'm saying this because I see ~ half of the scores in __no_feature and another half in __ambiguous. Try -s yes!

[lilingjoyo]

In fact, I tried both. They put almost same result. All reads are in "no feature" or "ambiguous" group.