[Biojava-l] Parsing Code for Phylogenetic Trees

[Michele Clamp]

On Wed, 14 Mar 2001, Emig, Robin wrote:

>	Speaking as someone who started with pre1.0 biojava, I think you may
>be changes with the following
>Sequences (these were introduced close to the 1.1 release)
>Gap symbols (I've been having problems with this implementation)
>Alignments (When you create an alignment, is doesn't keep the order of the
>sequences in the alignment because the current interface uses a
>java.util.Map)
>General parsing and converting to a String of symbol lists and symbols,
>sequences. Matthew has been talking about changing the way this between 1.1
>and 1.2.

Hmm Ok.  It seems as though we'll have to get stuck in if this is going to
happen.  As Matthew is in the office next door to me this shouldn't be
difficult :)

>	Having said that, even the changes that happened between
pre1.0
and >1.1 were not too code obliterating, since the use of interfaces is
>prevalent.

Yes I like interfaces very much.

>
>	However, It would be really great if you could mix and match
>components of JalView in Biojava. ie if you have ever seen Vector NTI and
>the way you can hide/rearrange the sequence view, the tree view, the plot
>view etc. I would love it if jalview panes were generic enough to say create
>an application which displays the alignment view and sequence logos, and the
>messaging is generic enough that they could talk to eachother (ie I select a
>position in the logo view and it selects it in the alignment view)
>	I don't want take Jalview out of your hands, but I would love it if
>all the code where in a cvs somewhere, where anyone who wanted to add to it
>could. Esp if it gets biojava-sized, having a bunch of biojava people able
>to add to it might be nice.


Sounds as though we're on the same wavelength here.  The new jalview will
have different panes that communicate specialized jalview events through
a central event controller.  Each pane reacts to those events as it sees
fit so an extra pane can be plugged in and if it knows about the events it
can happily coexist with everything else.  The basic events we have at the
moment are things like SelectionEvent, ViewEvent (scrolling basically),
EditEvent, RedrawEvent (for full redraws).  I don't know if biojava has
these sorts of events already. If so we could look at incorporating them.

As for the cvs Steve and myself are chopping away at cruft and trying to
get it to compile :) When that happens it will initially go into the same
cvs repository as apollo and ensembl and be available to all read only.  
If enough people are interested then it could be moved somewhere more
public.  

I am very happy for people to come in on the code but I have to keep in
mind that we need to be on top of the codebase here as at the end of the
day we want to use it in house for research purposes.  I have had tips of
fingers burned in the past with this (mostly due to lack of time to keep
track of things) and, of course, jalview is my baby and I don't want it to
leave home :)

M.


-- 
And so as the stripey-winged owl's genome of Fate 
is decoded by the great sequencer of Time,
and as the big grep of Eternity uses all the cpu of Destiny
I come to the end of the mail.