Hi,
I have to do a comparison among three groups of genetic transcription
levels.
I have a situation like this:
group 0: baseline
group 1: first treatment
group 2: second treatment
In the first group, I have only 2 samples, in the second one 4 samples and
in the last group I have 10 samples.
I would check if the trnascription of a gene increases from the baseline
situation to the fist treatment AND from the first treatment to the second
treatment.
I did a table like this:
GENE-LEVEL, GROUP
10,0
12,0
22,1
23,1
21,1
24,1
31,2
32,2
33,2
32,2
31,2
32,2
33,2
31,2
32,2
33,2
and I did this analysis:
lm(GENE-LEVEL~GROUP)
using the GROUP as integer and not as level.
Is it correct? and if the relationship is not strictly linear, ie the second
treatment level is not the double of the first treatment, is this analysis
still good?
Mainly I would only check whether group0<group1<group2 (or viceversa
group0>group1>group2)
Thank you
Michele Pinelli
------------------------------------------------------------------------
Michele Pinelli, MD, PhD
Complex Disease Genetics Unit
Department of Cellular and Molecular Biology and Pathology "L.
Califano"
University "Federico II", Naples
Italy
michele.pinelli@unina.it
Skype: dr.sson (Michele Pinelli/Naples, IT)
------------------------------------------------------------------------
"People still respect science and technology. … There is a problem however.
They have no idea what science is."
by Alan Leshner,
CEO of the American Association for the Advancement of Science
[[alternative HTML version deleted]]
