[BioC] Vsn implementation of spot weights

[w.huber at dkfz-heidelberg.de]

Hi Tom,

vsn does not use spot weights.

The purpose of vsn is not to make statements about individual genes, but
only to calculate a normalization and data transformation per array (or
per color channel). Typically, these are two parameters (an offset and a
scale factor) per array. A few outliers or bad spots among the thousands
will not make a difference, since vsn uses a robust estimation procedure
that will ignore such outlying data points anyway.

Subsequent per-gene hypothesis tests or regression procedures may benefit
from honoring spot weights, though.

Best regards
  Wolfgang

-------------------------------------
Wolfgang Huber
Division of Molecular Genome Analysis
German Cancer Research Center
Heidelberg, Germany
Phone: +49 6221 424709
Fax:   +49 6221 42524709
Http:  www.dkfz.de/mga/whuber
-------------------------------------

On Wed, 24 Sep 2003, Tom Price wrote:

> Does vsn implement the 'maW' spot weights?
>