htseq-count no hits

Question: htseq-count no hits

2.4 years ago by

VY • 120

London

VY • 120 wrote:

Hello,

I'm having quite abit of trouble handling my RNAseq fastq files. These were originally obtained from a single cell RNAseq experiment (probably not relevant). As usual I map back to mm10 using HISAT2 and then using a UCSC GTF I check for hits using htseq. The issue is that htseq doesn't seem to pick up any hits, it only does so in 1 or 2 genes.

I thought something might be wrong with my aligning so loaded the bam. file from hisat into IGV just to check where the reads fall and they do fall in genes, definitely more than 2 so im not sure what I'm doing wrong with my htseq settings. Any suggestions?

Thanks

rna-seq • 1.0k views

ADD COMMENT • link •

modified 2.4 years ago by Jennifer Hillman Jackson ♦ 25k • written 2.4 years ago by VY • 120

I've checked out this thread which was quite similar but it didn't seem to help (What to do when alignment rate is low even though the genomic data and RNA-seq data are of same stain)

This is the output I get from Hisat:

3337547 reads; of these: 3337547 (100.00%) were paired; of these: 3337528 (100.00%) aligned concordantly 0 times 12 (0.00%) aligned concordantly exactly 1 time 7 (0.00%) aligned concordantly >1 times ---- 3337528 pairs aligned conco

ADD REPLY • link written 2.4 years ago by VY • 120

2.4 years ago by

Jennifer Hillman Jackson ♦ 25k

United States

Jennifer Hillman Jackson ♦ 25k wrote:

Hello,

A few items to check:

All reference genomes are exactly the same between inputs. I am not sure where you mapped, but if not at http://usegalaxy.org, "mm10 reference genome" at Galaxy Main may or may not exactly match "mm10 extracted data" from UCSC. There could also just be a data entry problem (mm9 mixed in with mm10, or even more divergent by accident). Here is how to check your inputs: https://wiki.galaxyproject.org/Support#Reference_genomes
The GTF from the UCSC table browser might not contain all of the attributes that the tool is expecting for the Gene track selected. Examine the content versus the attributes the tool form states it needs. If any are missing, another source is needed.
Make certain the input BAM is "queryname" sorted, as described on the form. Samtools has a Sort tool. I would expect an error rather than this result if this was the actual problem, but it is critical for a successful run, so am mentioning it.
Examine other form settings to make sure alignments are not being ignored (for example, "stranded or unstranded" can make a big difference (even though having that specific option not match the data would produce slightly different results). Try defaults, then tune to make stricter rather than going in strict to begin with (if you were).

Hopefully one of these resolves the issue, Jen, Galaxy team

ADD COMMENT • link written 2.4 years ago by Jennifer Hillman Jackson ♦ 25k

Similar posts • Search »