[GeneX-dev] Re: CyberT

[Tony L. <td...@uc...>]

You could rewrite the code using some sort of power-law.  But I think 
most flavors of the power law would result in errors being Gaussian 
on a log scale....

At 8:12 AM -0700 5/22/01, Harry Mangalam wrote:
>Illes J FARKAS wrote:
>>
>>  On Mon, 21 May 2001, Harry Mangalam wrote:
>>
>>  > Date: Mon, 21 May 2001 20:08:12 -0700
>>  > From: Harry Mangalam <man...@ho...>
>>  > To: fi...@el...
>>  > Cc: wd...@nc..., genexdev at SF <gen...@li...>
>>  > Subject: Re: Cyber-T all-in-one
>>
>>  > Please let us know if you need help in installing the system.  I'll be
>>  on the road for the rest of the week, but the other GeneX developers will
>>  help. as long as you post to:
>>  >
>>  > gen...@li...
>>
>>  Thank you for the answer,
>>
>>  1) If I'm not mistaken, the Cyber-T software assumes that the distribution
>>  of data points (for one gene) is normal (i.e., Gaussian).
>
>As written, that's right.
>
>>  Or is it possible to use a power-law, for instance?
>
>If you wanted to use a power law, you'd have to re-write some of the 
>R library I think - Tony, could you comment?
>
>>  2) To use the paired data analysis page (for cy3/cy5) do I have to compute
>>  the Paired Expression Value Column? It seems to me that this column
>>  contains no information beyond the measured values, but it is simply a
>>  function of the measured control and data values.
>
>You have to compute this ONLY if you want the benefit of the 
>Bayesian analysis.
>It IS a function of the measured control and data values, but that 
>is what the Bayesian prior operates on.
>
>>
>>  Thank you for your help in advance.
>>
>>  Best,
>>  Illes Farkas
>>
>>  --
>>  Illes J FARKAS
>>  http://angel.elte.hu/~fij
>
>--
>Cheers,
>Harry
>
>Harry J Mangalam -- (949) 856 2847 (v&f) -- hj...@nc... || man...@ho...

-- 

Tony Long

Ecology and Evolutionary Biology
Steinhaus Hall
University of California at Irvine
Irvine, CA
92697-2525

Tel:  (949) 824-2562   (office)
Tel:  (949) 824-5994   (lab)
Fax: (949) 824-2181

email:  td...@uc...
http://hjmuller.bio.uci.edu/~labhome/