[dpryan]

It doesn't like the last line from these few lines (the lines before it are fine... this is actually the 74th line of my sam file)

Are those two reads sufficient to cause the crash, or are the earlier lines also required? Also, could you post the exact command you used that caused the crash (with options, like strandedness)?

[dpryan]

FYI, I was able to reproduce this crash in HTSeq-0.5.3 but it ran correctly in HTSeq-0.5.4. If you're not using the most up-to-date version, try upgrading and see if that fixes things.

[DrS]

Thank you - I'll try to do that. And yes, those were sufficient to make it crash.

[DrS]

Public License v3. Part of the 'HTSeq' framework, version 0.5.4p1.


Already running 0.5.4

[dpryan]

The most recent version is 0.5.4p3

[DrS]

Aha - on your website it only has up to p2, and the p2 was a Windows fix, where I'm on Linux. I'll download and let you know, but it may take a little while (working on other things now as well)

[dpryan]

Aha - on your website

I'm not the author, I just use htseq-count pretty frequently (I've never met the author!), so I'd like to get bugs fixed before they affect me too

[DrS]

I just downloaded and installed the one from:

https://pypi.python.org/pypi/HTSeq

The file is:
HTSeq-0.5.4p3.tar.gz

The version still reports as:
Written by Simon Anders ([email protected]), European Molecular Biology
Laboratory (EMBL). (c) 2010. Released under the terms of the GNU General
Public License v3. Part of the 'HTSeq' framework, version 0.5.4p1.


.... And I still get the crash.

100000 GFF lines processed.
200000 GFF lines processed.
296737 GFF lines processed.
Warning: Malformed SAM line: MRNM != '*' although flag bit &0x0008 set
Warning: Malformed SAM line: RNAME != '*' although flag bit &0x0004 set
Segmentation fault



I'm thinking I'm going to have to try another aligner and see if that works

[dpryan]

It's actually an easy fix to get the most recent version running. If you open the htseq-count file in a text editor, you'll see that it mentions the exact version (you can have multiple versions installed). Assuming everything with the newest version is setup correctly, you can just change "p1" to "p3" and things should work. BTW, which aligner are you using? I've had good experience with tophat2 and, more recently, STAR.

[DrS]

I was using Bowtie2 - paired-end genomic data from a 250x250 miSeq run.
I've asked others to chime in if they're also only having problems with paired-end data.

I'll try STAR next, but have meetings most of today so it might take a day or two to get to it.

And sorry about the mistaken identity

[dpryan]

Just keep in mind that you'll need rather more RAM for STAR, though it can finish a sample in the time it takes to grab a coffee.

[DrS]

RAM is not an issue in my group.

[DrS]

Just as an update - I ended up using STAR and had no problems with it. Still no idea why I was having problems with the other data.

[newbietonextgen]

GMAP and HTseq problems

I use GMAP to align my RNA-seq and then I use HT-Seq to get the counts. When i run the HTseq count script, it always complains of the sam file not being sorted.

I have used from picard tools, samtools and sort command to sort my SAM file but still it complains.

Picard tools (SortSAM)
SAmtools (convert from SAM to BAM, sort BAM file, Index BAM file, reconvert to BAM file).


Any reason? The HT-Seq count tools works well with tophat alignments