[maker-devel] A suspected strange behavior of Maker...

[Carson Holt]

For scaffold5347 I get results from exonerate and blastn without any
modification to default parameters.  For scaffold3928 the alignment is very
poor, and gets filtered out.  I can force it to keep it but I have to allow
for abnormally long introns of 65kb (introns that large are extremely rare
in biology ­ as in if you took every gene from 100 organisms you might find
a single gene among all of them with that long of an intron), and I have to
allow for low end to end matching (less 30%).  This alignment definitely
should have been rejected.

Thanks,
Carson


From:  Xin Huang <xh33 at georgetown.edu>
Date:  Monday, November 11, 2013 at 8:52 AM
To:  Carson Holt <carsonhh at gmail.com>
Cc:  <maker-devel at yandell-lab.org>
Subject:  Re: [maker-devel] A suspected strange behavior of Maker...

Hi Carson,

Thank you very much for getting back to me on this. Sorry I was too quick to
jump to a suspicion! Thanks for pointing it out and please see attached the
EST (face_cDNA.fa) and genome scaffold sequences
(albo_scaf3928+5347_for_face_cDNA.fa) I used for Maker. We used blastn to
identify these two scaffolds for this cDNA sequence, and exonerate result
also showed that this cDNA can be aligned to the scaffolds.

Thanks!

Xin 


On Mon, Nov 11, 2013 at 12:10 AM, Carson Holt <carsonhh at gmail.com> wrote:
> Sorry for the slow reply, I was moving from Canada to the US, got sick, and my
> moving truck got delayed one week. So I fell behind on the maker list.
> 
> What you observe is correct behavior.  The config is blasted against the
> database of ESTs/proteins.  It really doesn't matter which way you blast, you
> pick the direction that is most efficient for the question you are asking, and
> you only have to adjust the Z value for database size if you want alignment
> scores to be reproducible both ways.
> 
> Not getting results means they were filtered out for some reason (you don't
> even have blastn results in your output).  If you can send a test dataset I
> can take a look and tell you which filter.
> 
> Thanks,
> Carson
> 
> 
> From:  Xin Huang <xh33 at georgetown.edu>
> Date:  Tuesday, November 5, 2013 2:40 PM
> To:  <maker-devel at yandell-lab.org>
> Subject:  [maker-devel] A suspected strange behavior of Maker...
> 
> Hi,
> 
> I found a very bizarre thing when I ran Maker several times. The blastn
> research seems to reverse query and subject the way it's specified in the
> maker_opts.ctl file. When I put the scaffold sequence after genome= and the
> cDNA sequence after est=, the blastn result actually used the scaffold as
> query and the cDNA as subject. I realized that this sounds unreal, so I
> reversed the places for the two and the blastn result is normal with the cDNA
> as query and the scaffold as subject, but at the same time, the GFF file is of
> the cDNA sequence. I tried a few times, and got the same result. I deleted
> Maker (2.28) and reinstalled the beta version (2.30), still getting the same
> result.
> 
> My wild guess was that this might be the reason why the est2genome prediction
> using the EST sequence didn't go into the GFF3 file, as stated in a previous
> email to the Maker-devel. I couldn't figure out how to test this, though. I
> wonder if there's anything dramatically wrong that I've been doing to get such
> odd results... If you'd like me to send you any output files or control files,
> please let me know and I'll promptly pass those along.
> 
> Thank you very much for considering my email.
> 
> Sincerely,
> 
> Xin Huang
> _______________________________________________ maker-devel mailing list
> maker-devel at box290.bluehost.comhttp://box290.bluehost.com/mailman/listinfo/mak
> er-devel_yandell-lab.org



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://yandell-lab.org/pipermail/maker-devel_yandell-lab.org/attachments/20131113/047635c9/attachment-0003.html>