[Bioperl-l] seqboot and random seeds

[Jason Stajich]

We  use a different random seed each time
look at the code in
Bio::Tools::Run::Phylo::Phylip::SeqBoot

the _run subroutine is where all the action happens.

 # odd random seed
 my $rand = (2 * int(rand(10000)) + 1);

You can experiment by just hardoding it to something odd like 7 and
perhaps only generating a few replicates to see that you get the same
consensus tree each time.


You might also try and wait a little while before expecting an answer to
your questions as you posted the first question on Sunday and most people
have not had time to read or respond to the mail or try and replicate the
problem.  Shawn Hoon will hopefully have a chance to try things out as
well this week and get back to you if/why things aren't working how you
expect.


-jason

On Sun, 13 Jun 2004, june tantoolvesm wrote:

> Hi,
>
> I was just wondering if someone could please tell me
> how random seeds are handled in the SeqBoot module.
> For example, is there a way to find out what random
> number is being used, is it the same each time?
>
> Thank you.
>
> Best regards,
> June
>
> --- Joe Felsenstein <joe at gs.washington.edu> wrote:
> > Date: Sun, 13 Jun 2004 08:38:09 -0700
> > From: Joe Felsenstein <joe at gs.washington.edu>
> > To: june tantoolvesm <junet9876 at yahoo.com>
> > Subject: Re: bootstrapping related question
> >
> >
> > June Tantoolvesm --
> >
> > > I have automated a few of your programs into a
> > > pipeline using bioperl
> > > (seqboot->protdist->neighbor->consense). My
> > problem
> > > is, the final trees I get from when I run the
> > programs
> > > manually and from when I use bioperl, are not the
> > > same. I tried several times.
> > >
> > > Is it normal to not get the same tree everytime we
> > run
> > > the pipeline of programs
> > > (seqboot->protdist->neighbor->consense). (But
> > actually
> > > I do get the same tree when I run the programs
> > > manually a few times, so I'm confused.) Also, does
> > > this have anything to do with the random seed
> > number
> > > we input for seqboot?
> >
> > You should get the same tree each time when you run
> > the
> > Bioperl script, provided that it properly renames
> > files,
> > and provided that the random number seeds used are
> > the
> > same each time.  There is no reason to expect them
> > to
> > be different.
> >
> > Check on the issue of random number seeds, and
> > whether the
> > output files have been renamed to input files
> > properly.
> >
> > J.F.
> > ----
> > Joe Felsenstein         joe at gs.washington.edu
> >  Department of Genome Sciences and Department of
> > Biology,
> >  University of Washington, Box 357730, Seattle, WA
> > 98195-7730 USA
> >
>
>
>
>
>
> __________________________________
> Do you Yahoo!?
> Friends.  Fun.  Try the all-new Yahoo! Messenger.
> http://messenger.yahoo.com/
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>

--
Jason Stajich
Duke University
jason at cgt.mc.duke.edu