Question: Design matrix for limma
0
gravatar for hakimelakhrass
4.0 years ago by
hakimelakhrass80 wrote:

Hello, I am new to microarray data and I am looking to make a design matrix. Following the limma documentation I made the following design matrix:

control controlvstreatment
 [1,]       1                  0
 [2,]       1                  0
 [3,]       1                  0
 [4,]       1                  0
 [5,]       1                  0
 [6,]       1                  0
 [7,]       1                  0
 [8,]       1                  0
 [9,]       1                  0
[10,]       1                  0
[11,]       1                  0
[12,]       1                  0
[13,]       1                  0
[14,]       1                  1
[15,]       1                  1
[16,]       1                  1
[17,]       1                  1
[18,]       1                  1
[19,]       1                  1
[20,]       1                  1
[21,]       1                  1
[22,]       1                  1
[23,]       1                  1
[24,]       1                  1
[25,]       1                  1
[26,]       1                  1

 

my results came out like this :

  control controlvstreatment
-1       0                517
0        0              25892
1    26806                397


Which look a little weird. Now I need to expand to 5 time points and I am not sure how. Thanks in advanced.

limma microarray hugene20st • 1.6k views
ADD COMMENTlink written 4.0 years ago by hakimelakhrass80

I'm not sure what looks so weird about those results. The "control" column is the intercept, so you can ignore it. How to deal with multiple time points depends on the question(s) you want to ask.

ADD REPLYlink written 4.0 years ago by Devon Ryan94k

Ah I see now. I am trying to find the DEGS between all time points. I am not sure exactly how to go about it. I have a baseline two time points in the middle and a time point that acts as an "after".

ADD REPLYlink written 4.0 years ago by hakimelakhrass80
1

So you presumably want most of the pairwise comparisons. Generally you would use a model like ~0+timepoint and use a bunch of contrasts, one for each comparisons you want to make.

ADD REPLYlink modified 5 weeks ago by RamRS25k • written 4.0 years ago by Devon Ryan94k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1664 users visited in the last hour