[BioC] loess and limma

[john seers (IFR)]

Hi Wolfgang

> so I am often also perplexed how people use such a
>heavy gun without much thought.


I am trying to avoid being one of those people. :)


John
 
---

John Seers
Institute of Food Research
Norwich Research Park
Colney
Norwich
NR4 7UA
 

tel +44 (0)1603 251497
fax +44 (0)1603 507723
e-mail john.seers at bbsrc.ac.uk                         
e-disclaimer at http://www.ifr.ac.uk/edisclaimer/ 
 
Web sites:

www.ifr.ac.uk   
www.foodandhealthnetwork.com


-----Original Message-----
From: Wolfgang Huber [mailto:huber at ebi.ac.uk] 
Sent: 14 September 2006 15:58
To: john seers (IFR)
Cc: sdavis2 at mail.nih.gov; bioconductor at stat.math.ethz.ch
Subject: Re: [BioC] loess and limma


Hi John,

> This worries me as people routinely slap in a loess normalisation. 

Non-parametric smoothing (such as loess) is an enormously powerful tool,
but getting it right (especially at the boundaries) can be tricky, and
there are parameters to choose such as the 'span' and the order of the
local polynomials, so I am often also perplexed how people use such a
heavy gun without much thought.

There is some potential for over-fitting and smoothing out real signal.

Well, that's my little rant for the afternoon :)

 Best wishes
 Wolfgang

------------------------------------------------------------------
Wolfgang Huber  EBI/EMBL  Cambridge UK  http://www.ebi.ac.uk/huber