I downloaded a GSE file from GEO and converted it to an expression-set using GEOquery. However, even though the values are normalized, they are not log-transformed which is kind of bad, since I'm using limma to analyse the samples.
Is it possible to log-transform the values of an eSet directly (without using exprs()), or can I tell limma to log-transform the values while using fit() or topTable()?
what platform is your dataset of interest ? and also what specific functions you have used? It is a bit weird you mention that the values are normalized but not log-transformed (but of course it depends on the normalization methodology).
Anyway, personally i think the simplest way is to download the raw data by using the function:
I'm using the Illumina GPL8432 platform. The Data is quantile-normalized but not log2-transformed.
As mentioned above, I manually log2-transformed the normalized values in order to use limma. Is this a bad thing?
Hello svlachavas,
I'm using the Illumina GPL8432 platform. The Data is quantile-normalized but not log2-transformed.
As mentioned above, I manually log2-transformed the normalized values in order to use limma. Is this a bad thing?
Thanks for taking your time,
M