How to inteprate distances calculated by cmdscale?
0
0
Entering edit mode
Adam • 0
@adam-10025
Last seen 4 months ago
Poland

Hello,


I have theoretical questions. I have differential gene expression data with values from 0 to 30000. I've calculated euclidean distances between probes in R. Mirror euclidean distance array has values between 0 and 5. Then I've used cmdscale method to gain points(probes) coordinates. Then I plot my results, genes are visibly grouped. My questions are:
1. Cmdscale method result is log2-fold-change?
2. Axes on plot: Y: from -1 to 2, X: from -2 to 2, I've measured distance between two points (by euclidean formula), the distance is 1.2, what this tells me? I mean difference between expression of genes.

 

Regards

Adam

differential gene expression cmdscale mds • 1.5k views
ADD COMMENT
1
Entering edit mode

It's not clear from your description what you have done. Giving a self-contained code snippet is probably better than just saying what you have done, because it's often the case that people have done something different than what they thought they did. What does 'Mirror euclidean distance array' mean?

Are you doing MDS on the samples, or the genes? MDS on the genes doesn't really make sense, but when you say 'genes are visibly grouped', the assumption for me at least is that you have done MDS on the genes.

I should also point out that cmdscale is a base R function, not Bioconductor, so you should probably be asking these questions on R-help (r-help@r-project.org). But you may well get flamed or ignored there because this is a statistics question, not a 'how do I do this thing using R' type question.

To that end, you might want to look at what Wikipedia has to say about MDS. Perhaps you can answer your questions yourself.

ADD REPLY

Login before adding your answer.

Traffic: 1034 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6