CLASS-L Archives

October 2006


Options: Use Monospaced Font
Show HTML Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Yakir Gagnon <[log in to unmask]>
Reply To:
Classification, clustering, and phylogeny estimation
Sat, 28 Oct 2006 20:28:30 +0200
text/plain (1346 bytes) , text/html (1549 bytes)
Hi all!
first timer here.
I need to find out to what degree (in the form of a P value or r^2) some
spectra are similar to each other or different. here is the problem:

   1. I have 10 groups; 10 taxonomic different groups of zooplankton
   (small marine animals).
   2. each group has between 4 and 11 replicates (I measured 4-11
   individuals from each specie).
   3. each replicate is a spectrum; for each replicate the spectrum is
   composed of a series of 600 values. 600 light intensity values corresponding
   to the 600 wavelengths they were measured at. the wavelength spans from 400
   to 700 nm.
   4. I want to know to what degree are these spectra similar to each
   other within the groups, and between the groups.

From eyeballing them I can say that they are pretty similar within the
groups in some groups and not so similar in other groups. I can also say
that they look rather different from each other between the groups. but I
need to know how significant it really is.
what is the statistically correct thing to do?
thanks in advance!

Yakir L. Gagnon, PhD student
The Vision Group
Tel  +46 (046) 222 93 40
Cell +46 (073) 7536354
Fax +46 (046) 222 44 25

CLASS-L list.