[Biopython-dev] Lowess Smooth Improvement

[Peter]

On Thu, Nov 13, 2008 at 1:00 PM, James Pruitt <eric.pruitt at gmail.com> wrote:
> I made some changes to the Lowess smoothing method as well as written a unit
> test for it. On my machine, it runs around 37% faster in my unit tests
> compared to the original lowess method and that is using the numpy.median
> function so it would probably run even faster with the Bio.Cluster median
> functoin.

Presumable this is an update for Bio/Statistics/lowess.py?  I'm a
little confused - this code already uses Bio.Cluster.median if it can,
falling back on numpy.median.  Maybe you're working from an older
version of Bipython?

> How do I go about proposing my code to be included in Bio.Python?

First file an enhancement Bug, then once the bug is filed you can
attached a patch against CVS.
If you have any example scripts or unit tests to go with it, even better.

Thanks,

Peter