Proposal for creation of algorithm repository

Malay Kumar Basu curiouser at ccmb.res.in
Fri Mar 21 22:07:07 EST 2003


> What you are proposing sounds like a VERY large task.  There are
> several open-source projects (BioPerl, BioPython, ... ) that have
> started doing things like this, and none of them have managed to do
> cover more than a small fraction of the important algorithms.
> Perhaps the best thing to do is to join one of the open-source
> projects and contribute a few of your favorite algorithms implemented
> efficiently.
> 

I know it's a huge task. I never said that I want do it myself. All I
wanted is input form the people whether this sort of repository is worth
creating or not. I told about a wiki like interface whether people will
submit code and others will correct or modify them. The whole process
should be a community driven venture. A knowledgebase for all.

Neither BioPerl or BioJava concentrate on the algorithmic aspect in
computaional biology. The majority of the code in both these projects are
just conversion of file formats, database hacking and such. I know I risk
flames by telling that. But my comment is just a matter of fact
observation. I should mention here I am not trying to demean those
projects. I am a great fan of both these projects and learnt a lot from
both.

With my limited knowledge in the field I humbly believe that the 
number of fundamental algorithms in computational biology are not so
profuse that it can not be handled by handful of dedicated people. Someone
just need to start the role of "chinese old man" and start cutting the
mountain. If the destination matches then other people will join. And
history says that there is no dearth of "old men" or dedication in
software hackers. After all that's "why I am (trying to be) a
bioinformatician"!

Best regards,

Malay 






More information about the Comp-bio mailing list