From: Barbier de Reuille <barbier@...>
Date: 2005-09-01T15:46:24+09:00
Subject: Re: Good code examples

Dans l'article <Pine.LNX.4.62.0508311740230.29620@harp.ngdc.noaa.gov>,
Ara.T.Howard a �crit :
> On Thu, 1 Sep 2005, Barbier de Reuille wrote:
> 
>> Thanks, that was exactly the kind of ressources I was looking for ! I'm very
>> interested in numerical computing, but not only ^_^
> 
> great - you'll find ruby a treat then.
> 
>> My main interest is on structural computing (applied to biology) but it uses
>> a lot of numerical computation ;)
> 
> can you elaborate?  i'm not biologist ;-)

Well, I'm working on some tissue growth. Tissues are here represented as
graph with nodes being cells and edges beeing the neighborhood
relationship. Computing the evoluation of this graph is largely
symbolic. However, you need to solve some ODE on this graph. Most of the
case these ODEs rule the concentrations of different proteins. Solving
these ODEs involve numerical computation.

Also, to study the structure of real tissues we extract quantitative
data from graph constructed using existing tissues. Once extracted we
can get some hundred or thousand data. We need statistical tools to
study them. And they're not easy (cyclic, spatial, ...) so we need tools
like GNU R and the big number of statistic libraries available.

> 
>> By the way, does anybody knows if there is some kind of GNU R ruby binding ?
>> If not it may be a good project as I think I can begin easily and improve it
>> slowly ...
> 
> http://raa.ruby-lang.org/project/ruby-rmathlib/
> 
> but i have not used it
> 
> -a

Pierre

-- 
BOFH excuse #431:

Borg implants are failing