cuffdiff complains SAM file being not sorted, when I am sure they are right.

[thejustpark]

Hello all,

I am running cuffdiff with different gtf files but with the same set of sam files.
For example, they look like this.
cuffdiff -b hg19.fa --max-bundle-frags 4000000 -o lincRNAs lincRNAs.combined.gtf a.sam b.sam c.sam
cuffdiff -b hg19.fa --max-bundle-frags 4000000 -o codingGenes coding.txt.gtf a.sam b.sam c.sam, where a.sam, b.sam, and c.sam are from different tissues.

In running them, I have successfully run the latter, but I got the following error for the former:
--------------------------------------------------------------------------------
[08:24:24] Loading reference annotation.
[08:24:24] Inspecting maps and determining fragment length distributions.

Error: this SAM file doesn't appear to be correctly sorted!
current hit is at chrX:61774, last one was at chrM:16495
Cufflinks requires that if your file has SQ records in
the SAM header that they appear in the same order as the chromosomes names
in the alignments.
If there are no SQ records in the header, or if the header is missing,
the alignments must be sorted lexicographically by chromsome
name and by position.
--------------------------------------------------------------------------------

But the thing is sam files cannot be wrong for the former running, because I am using the same set of the sam files. As it happens in loading reference annotation, I am thinking this might be related to lincRNAs.combined.gtf in the example, but actually this file is a result of cuffcompare (as reflected in the name), another package in cuff serires tools. So, I cannot think this file is incompatible with cuffdiff.

I am wondering where I have to look into to address this issue.

Thanks a lot for your advices in advance.
HJ.

[swbarnes2]

The error message sounds pretty straightforward, is it incorrect about the order of your headers?

Try using Picard to sort, if you aren't already.