Question

design matrix for a 2 groups

0

Entering edit mode

AAAbdul • 0

@aaabdul-10719

Last seen 7.2 years ago

My experiment consists of 2 groups (54 non-diabetic and 9 diabetic). I would like to use limma to find the differentially expressed genes for these microarray data set.

http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE38642+

I did Background correcting, Normalizing and Calculating Expression, unfortunately, I do not know how to create design matrix to use limfit() function

I am a beginner in R so any help is appreciated.

limma • 870 views

ADD COMMENT • link updated 7.9 years ago by chris86 ▴ 420 • written 7.9 years ago by AAAbdul • 0

score 1 · Answer 1 · 2016-05-17

1

Entering edit mode

chris86 ▴ 420

@chris86-8408

Last seen 4.4 years ago

UCL, United Kingdom

This is really easy, I suggest you consult the limma manual https://www.bioconductor.org/packages/3.3/bioc/vignettes/limma/inst/doc/usersguide.pdf. You use the model.matrix function to build the design matrix for your model. The order of the model matrix has to match your data column/sample order.

design <- model.matrix(~group) # where group simply is a vector of status e.g. 0 for no disease, 1 for healthy
matrix <- data.matrix(data)

fit <- lmFit(matrix,design)

If you still cannot get your head around it, it may be best to find some one that can.

ADD COMMENT • link 7.9 years ago chris86 ▴ 420

0

Entering edit mode

Could you explain this a little bit @Chirs86 please?

how can I create groups vector ?

the samples not in sequence it arranges like 2 disease and 12 healthy then 3 disease and 25 healthy.

as you see it comes like this

status: non-diabetic donors

status: T2D donors

status: non-diabetic donors

status: T2D donors

status: non-diabetic donors

status: T2D donors

status: non-diabetic donors

status: T2D donors

status: non-diabetic donors

I did like this

strain <-c("ID1","ID2","ID3","ID4","ID5","ID6","ID7","ID8","ID9","ID10","ID11","ID12","ID13","ID14","ID15","ID16","ID17","ID18","ID19","ID20","ID21","ID22","ID23","ID24","ID25","ID26","ID27","ID28","ID29","ID30","ID31","ID32","ID33","ID34","ID35","ID36","ID37","ID38","ID39","ID40","ID41","ID42","ID43","ID44","ID45","ID46","ID47","ID48","ID49","ID50","ID51","ID52","ID53","ID54","ID55","ID56","ID57","ID58","ID59","ID60","ID61","ID62","ID63")
> design <- model.matrix(~factor(strain))

but I do not know how to create colnames(design)

I am a beginner in R and I will be greatful to you

ADD REPLY • link 7.9 years ago AAAbdul • 0

0

Entering edit mode

I normally would have an annotation file, which is a data frame. In this there would be column 1 = your ids in the same order as microarray (google for how to make data frames and reorder columns of matrix and dfs) column 2 = disease status and then other columns like your covariates like gender and age. Then your model matrix is simply model.matrix(~factor(des$status)). Where des is the description file.