[EGrassi]

Hi EGrassi,

Did you solve this issue??????

Actually no, right now I'm trying to avoid using '-' as an htseq-count argument without using it in a pipeline with samtools view, I'll let you know what happens...

[Simon Anders]

Woa, I used samtools -n and htseq-count gave me all counts of zero...that's strange, I will check with IGV or similar tools but cuffdiff gave me fpkg values with the same gtf and bam file...

Was this by any chance a TopHat output? Somebody alerted me recently to a bug that causes mishandling of SAM lines that contain the "NH" flag, and that seems to appear in newer tophat output. I'll try to fix this ASAP.

Simon

[EGrassi]

Was this by any chance a TopHat output? Somebody alerted me recently to a bug that causes mishandling of SAM lines that contain the "NH" flag, and that seems to appear in newer tophat output. I'll try to fix this ASAP.

Tophat 2.0 output, yes!

Thank you,
E.

[EGrassi]

Yep, I can confirm you that removing the NH flags from a subset of my sam files I get some counts different from 0. This could be a temporary workaroud.

[EGrassi]

Ok, on the whole data it falls back to all 0:
samtools view S1.sorted.bam.bam | sed 's/NM:i:\d//g' | htseq-count --stranded=no - /rogue/bioinfotree/prj/ewing-rnaseq/local/share/data/Homo_sapiens/UCSC/hg19/Annotation/Genes/genes.gtf > counts_S1_htseqq

I'm really a little surprised by the amount of errors and other glitches in rna seq software (this is a generic rant, not focused on htseq!)...it's a really complicated and new area, that's true, but right now using any of the existing tool seems a bet and to tell the truth I never feel confident on the results, as long as with every version results changes, and when something runs without errors I never know if that means that it worked or that an exception just had not screamed enough :/

[Simon Anders]

The bug with the NH flags, which was introduced during some refactoring in the version of a week ago or so is now fixed in version 0.5.3p8. Sorry that the recent changes turned out to be so volatile.

[EGrassi]

Thanks, 0.5.3p9 has produced counts different from zero. Let's see what deSeq thinks of this data!

[sushant]

Sorry, it seems we made some mix-up between version 0.5.3p4 and 0.5.3p5. Essentially, p5 undid some fixes in p4, including the one for "*" qualities. Now, there is version 0.5.3p6, which should clean up this mess. Please let me know if you still have problems.

Hello, I have performed the mapping of miRNA reads to the mirbase precursor file by SHRiMP. But I'm facing a probem in HTSeq while taking the read counts from the output.sam file.

The error which I'm trying to solve is;

Error occured when processing SAM input (line 2305 of file HBI_10.sam): ("'seq' and 'qualstr' do not have the same length.", 'line 2305 of file HBI_10.sam') [Exception type: ValueError, raised in _HTSeq.pyx:808]

Has anyone of you come across this error? What might be the solution?

[maubp]

Sushant - what does line 2305 of your SAM file look like? Unless the quality entry is the special value '*' only then it is a likely an unrelated issue.

[sushant]

Sushant - what does line 2305 of your SAM file look like? Unless the quality entry is the special value '*' only then it is a likely an unrelated issue.

Actually, I did see that SAM file. The problem is that my sequence string & its respective sequence quality string is not of equal length.
However, I converted my fastq files to fasta and then performed the alignment, which solved this problem.

But the mapping percentage I'm getting with SHRiMP is too low(2.77%), for the mirna aligning to mature.
What parameters should I consider?