Hi all,
I am trying to perform an RNAseq analysis. I have created an index using HISAT2, aligned my paired-end reads to the index, and am now trying to count reads using HTseq.
Unfortunately, when counting, HTseq yields the following result.
__no_feature 6280013
__ambiguous 0
__too_low_aQual 4434361
__not_aligned 4758201
__alignment_not_unique 3366408
What I have done to troubleshoot this.
- Create new index using different file (transcriptome instead of genome)
- Use .gff or .gtf file for counting.
- Sort .sam file prior to counting.
- Ensure
--stranded=noand--nonunique=all
These changes have not yielded correctly counting output.
At this point I am not sure what to do. I imagine there is something wrong with the way I set my index up and potentially not matching the .gtf file, however, I have downloaded these from the same source so I am not sure.
Thanks
1 answer
I dont see chromosome names in my .gtf file, and I thought this would not be a concern as I am using gene ID as the ID attribute (not position)? Is that not correct?
Im not sure how to attach the file itself but here is a screenshot of the .gtf and the first lines of the .sam file after I aligned...
It appears that you have aligned the data to a transcriptome file and not the genome. XM_* accession number and the length show that. So this is not going to work.
You will need to align your data to the genome NC_037638 and then count using the GTF.
If you want to use the transcripts to quantitate against then you will need to use a program like salmon instead.
Thank you for your help. I appreciate it.
Log in to answer this question.
-
what is the normal range of htseq-count output for homo sapiens RNA-Seq
written by Morteza •
-
Htseq-count output file having a high number of __not_aligned
written by Nemo •
-
htseq count error
written by Balatheskulter •
-
Should I keep all HTSeq table informations for edgeR analysis?
written by silas008
-
Ambiguous reads in htseq-count
written by kousi31 •
-
strange htseq-count result
written by lilingjoyo •
-
Htseq-count Why Do I have more nofeature readswhen i'm changing nonunique option ?
written by bjmnpnd •
-
stranded vs not stranded Htseq-output differences
written by AP •
-
HTseq output questions
written by sophialovechan •
-
Does better results in HTSeq-Count mean that the script was run correctly?
written by cr.balladares.m •
Have you examined the alignments with IGV or a similar genome browser? Are the reads aligning in correct place (e.g. under exons). Also try using
featureCountsto see if the result is any different.Yes, I have looked at them in a genome browser and they look ok, so I am confused why HTSeq is unable to count them. I tried using featureCounts per your suggestion but this has resulted in 0 reads being counted.