HyPhy message board
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl
HYPHY Package >> HyPhy bugs >> DirectionalREL.bf
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?num=1225247063

Message started by walter on Oct 28th, 2008 at 7:24pm

Title: DirectionalREL.bf
Post by walter on Oct 28th, 2008 at 7:24pm
Sergei,

I have used HyPhy for quite a while. I do not write batch files though. Simply use what you provide :). Never had a problem like one I am writing about. When trying to execute DirectionalREL.bf, see the following messages (see the pictures). What's wrong? I use the same type of the data files as before: NEXUS or FASTA.

Untitled.jpg (44 KB | )

Title: Re: DirectionalREL.bf
Post by Sergei on Oct 28th, 2008 at 9:51pm
Dear walter,

DirectionalREL.bf expects a protein (not a codon) alignment. Could you verify that's what you were inputting?

Cheers,
Sergei

Title: Re: DirectionalREL.bf
Post by walter on Oct 29th, 2008 at 12:48am
Of course, I used an amino acid alignment. I even generated a nexus file from HyPhy and then used it as an input.

Title: Re: DirectionalREL.bf
Post by Sergei on Oct 29th, 2008 at 1:34pm
Dear Walter,

Could you attach (or e-mail to me) the alignment file that causes DirectionalREL.bf to throw an error message so that I can reproduce the error and try to debug it? The error report suggests that HyPhy is reading your alignment as something other that proteins (this is why the incompatible dimensions error), and I am not sure why.

Also, which analysis options did you select for DirectionalREL.bf?

Cheers,
Sergei

Title: Re: DirectionalREL.bf
Post by walter on Oct 29th, 2008 at 8:06pm
Sergei,

The file is attached. Also, please note that there are similar problems with some other analyses (e.g., amino acid model test), although I have not tried all. One more: when I close the program window, I see a message: Program stops working.
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?action=downloadfile;file=cikA_aa_trimmed.bf (83 KB | )

Title: Re: DirectionalREL.bf
Post by Sergei on Oct 29th, 2008 at 9:47pm
Dear Walter,

Hmm; looks like when you save a protein file from the HyPhy datapanel, HyPhy incorrectly generates the NEXUS header by claiming this to be a DNA datafile, e.g

[code]
FORMAT
           DATATYPE = DNA

           GAP=-
           MISSING=?
     ;
[/code]

when of course DATATYPE=PROTEIN should be used. Thanks for pointing out this erroneous HyPhy behavior; I have fixed it in our development code already (this will be included in the next s/w release). If you edit the NEXUS file and change DNA to PROTEIN, the problem will disappear.

Before the next version of code is released, I would suggest doing this to avoid the issue: in the datapanel, make a partition using the entire dataset, and then click on the 'Save Partition To Disk' button to export the file (see sections 2.3.2 and 2.3.3 in Multimedia File Viewing and Clickable Links are available for Registered Members only!!  You need to Login Login)

Cheers,
Sergei  

HyPhy message board » Powered by YaBB 2.5.2!
YaBB Forum Software © 2000-2024. All Rights Reserved.