Htseq question

[fanli]

A (sort-of) workaround would be to count ambiguous reads for both genes:

http://seqanswers.com/forums/showpost.php?p=161401&postcount=7

[dpryan]

This seems to be an interesting edge-case. So, TUBB3 and MC1R can form fusion transcripts and fusion proteins. So ENSG00000198211 isn't a mistake and htseq-count isn't doing anything incorrect. However, cases like this are really annoying for read counting. The simplest solution would be to just remove the entries for ENSG00000198211 in your GTF file. The actual correct solution would require a different annotation format that can account for non-novel fusion events of genes that are properly annotated and analyzed as separate in many cases. Such a format will probably get created someday, but I imagine it'll take a while.

[zjrouc]

Just use the -o option to track how the reads are being categorized. Then you can simply look and see why they're not being counted.

Dear dpryan,

Thank you for your reply. I have checked the results, It showed htseq classified most of the reads of this region into "ambiguous" condition, and they belong to ENSG00000198211.8 + ENSG00000258947.3. However, i checked these two genes. Both of them are TUBB3 gene. One is Chromosome 16: 89,921,392-89,938,761 and the other is Chromosome 16: 89,919,165-89,936,092. I understand this problem is not caused by the Htseq software. However, if we discard these reads, it will have a huge impact on expression of this TUBB3 gene (Htseq count 7, but there are dozens).

Do you have any ideas about this?


Thanks

[dpryan]

Just use the -o option to track how the reads are being categorized. Then you can simply look and see why they're not being counted.

[zjrouc]

Hi, all,

i just got a result from ht-seq. It showed that my interesting gene has 7 counts in the alignment. However, from the view of IGV, i could easily identify much more counts than 7 on this gene. My alignment is from STAR, and i used more stringent parameters to control the multiple alignment, which means there should not be any multiple aligned reads in the output. I am really confused about this.

Any suggestion?