----- Forwarded message from "Schmidt, Eric" -----
Date: Mon, 2 Jun 2014 16:34:35 -0400
From: "Schmidt, Eric"
Reply-To: "Schmidt, Eric"
Subject: cluster analysis advice
To: "morphmet@morphometrics.org"
I'd like to ask for some advice on using kinds of cluster analysis for a
morphological data set. Shape is not involved. I have data on the weights of
certain organs from numerous mice. These included things like livers, spleens,
hearts, and kidneys, dissected from adult mice. One aim is to take the data
and find modular patterns within them, without making any a priori hypotheses.
I'd also do this with the body weight residuals of all the organs.
One method I will try is to do a PCA, using organ types as the variables. If
different organ types tend to group and cluster, perhaps those "clusters" could
be used as a priori hypotheses in a PLS regression. Another way to group
organs might be to base clusters on germ layers, and do PLS. However, I'd
like some advice on a method or set of methods that will do a cluster analysis
independent of an initial hypothesis.
Thank you
Eric
Wash U
-------------------------
The materials in this email are private and may contain Protected Health
Information. If you are not the intended recipient be advised that any
unauthorized use, disclosure, copying, distribution or the taking of any action
in reliance on the contents of this information is strictly prohibited. If you
have received this email in error, please immediately notify the sender via
return email.
----- End forwarded message -----
