This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Get Expression Table from CEL raw data with R

Hi everybody, I am a newbie working with R, so please excuse some potential confusion.

I have some CEL raw files and would like to generate a table with expression levels. My CEL data is based on a Clariom S Mouse CHIP.

I used the oligo package to import and read my CEL files, which gives my a Large ExpressionFeatureSet. I created a matrix using: 

exp_raw <- log2(Biobase::exprs(raw_data))

For the obtained matrix I get unspecified features (just numbers starting from 1...). I would like to assign the features to the Clariom S Mouse data CHIP that was used (meaning the gene names from the CHIP). Is there an easy way to do that?

In addition to that I can do some simple quality analysis but I don’t know how I can define groups for my samples, to compare them (I have 6 samples and two different genotypes that I want to compare). Do you have a recommendation how to deal with that? Do I need to create some Array Design Format (ADF) or Sample and Data Relationship Format (SDRF)?

Thank you very much in advance! I am happy for every advice!

cel microarray r
• 3,942 views • link
written by lsbw • lsbw

Dear Kevin,

thank you very much for your answer and please excuse my late response! I really appreciate you trying to help me!

So I used the following steps:

first I set my working directory

next I load oligo

library(oligo)

I read my CEL files using

raw_data <- read.celfiles(list.celfiles())

I create a matrix from my expression set (log2 to get the fold change)

exp_raw <- log2(Biobase::exprs(raw_data))

Concerning the annotation problem, I will try to follow the instructions from the link you send. Already thanks a lot! Lisa

• 0 views • link
written by lsbw • lsbw

Hey, that looks okay. However, are you sure that you want to be using the log2 of the raw intensities?

If you want to normalise your data, then should could do:

library(oligo)
raw <- read.celfiles(list.celfiles())
normalised <- rma(raw)

Then, you can access the log2 normalised expression levels via:

exprs(normalsied)

Try:

boxplot(exprs(normalised))
hist(exprs(normalised))
• 0 views • link
written by Kevin Blighe Kevin Blighe

1 answer

Hey isbw, and welcome / willkommen.

So that I can be sure that you are using the correct data, can you please show the steps that you used for reading in the CEL files, and also your oligo (and other steps)?

Also, this following command may not be doing what you expect:

exp_raw <- log2(Biobase::exprs(raw_data))

Can you elaborate on what you think is happening with this command? If you have used oligo, then your ExpressionSet object should already contain normalised, log2-transformed, expression levels. Showing all of your commands will help to determine what is happening, though.

--------------------

In terms of annotation, you can use the pre-built Bioconductor annotation package, pd.clariom.s.mouse. To get an idea of how to use this type of annotation package, take a look at my answer here (see the third code-chunk): https://support.bioconductor.org/p/130727/

Kevin

• 0 views • link
written by Kevin Blighe Kevin Blighe

Log in to answer this question.

More posts like this