Entering edit mode
Paolo Kunderfranco
▴
350
@paolo-kunderfranco-5158
Last seen 7.2 years ago
Dear All,
I am new to multifactor analysis with edgeR, I have the following 12 samples (4 conditions and 3 replictaes)
| sample | type | diet |
|---|---|---|
| sample1 | wt | no |
| sample2 | wt | no |
| sample3 | wt | no |
| sample4 | ko | no |
| sample5 | ko | no |
| sample6 | ko | no |
| sample7 | wt | yes |
| sample8 | wt | yes |
| sample9 | wt | yes |
| sample10 | ko | yes |
| sample11 | ko | yes |
| sample12 | ko | yes |
I would like to find genes with these properties:
- genes differentially modulated between wt-no_diet group (sample1-3) and ko-no_diet group (sample4-6)
- genes that are affected by the diet treatment only in ko group
- genes whose expression in the ko-yes_diet group (sample10-12) tends to get similar to the expression in the wt-no_diet group (sample1-3).
Any suggestion will be greatly appreciated,
Best
Paolo
Dear Aaron,
I formatted the table as requested, now it should be much more readable. I read in the user-guide but I find it difficult to create this multi-factor analysis,
Best
Paolo
Read chapter 3, and section 3.3.1 in particular. Just combine the two factors into a single factor and set up the design matrix with four groups, i.e., one group for each combination of diet and genotype.