Hi Dave,
Thanks for the heads up!  Definitely a good idea for us to incorporate that
into our X!Tandem distribution (will do so for our next release); perhaps
also a good idea to report it to Ron?
Cheers,
--Luis



On Mon, Oct 17, 2011 at 9:26 AM, [email protected] <
[email protected]> wrote:

> Hi all,
>
> I've come across problems running X!Tandem searches against NCBI nr
> derived FASTA files due to some entries having extremely long header
> lines.
>
> Current code in msequenceserver.cpp is:
>
> 166     // 2006.11.21 - increased the size from 32*4096 to 512*1024
> because of very long lines in nr FASTA files
> 167     m_lSize = 512*1024-1;
>
> The longest line in a human extract from nr is now above 512KB:
>
> [manager@proteomics-srv-01 ncbi_nr]$ cat decoy_nr_homo_sapiens.fasta |
> awk '{print length}'|sort -nr|head -1
> 684322
>
> ... so that's ~684KB, exceeding the 512KB in msequencesserver.cpp.
> This is causing a segfault when searching against nr.
>
> I've increased to m_lSize = 1024*1024-1 and am just wondering whether
> this should be done in the TPP distribution, as presumably many people
> could come up against this segfault issue if they use nr. Of course,
> 1MB for a FASTA sequence header is a bit ridiculous!
>
> Cheers,
>
> DT
>
> --
> You received this message because you are subscribed to the Google Groups
> "spctools-discuss" group.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to
> [email protected].
> For more options, visit this group at
> http://groups.google.com/group/spctools-discuss?hl=en.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"spctools-discuss" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/spctools-discuss?hl=en.

Reply via email to