Question

how/what to use to perform Differential Expression Gene if log2 available?

0

Entering edit mode

Michael • 0

@michael-10785

Last seen 9.5 years ago

Hi,

I am trying to use TCGA level 3 microarray data to get DEGs. The files first columns are genes and second columns are "log2 lowess normalised". Can I use this to get DEGs? if yes, what package can use that normalised log2 lowess value for this task?

Thank you very much

microarray limma edgeR • 929 views

ADD COMMENT • link updated 9.5 years ago by Gordon Smyth 53k • written 9.5 years ago by Michael • 0

score 1 · Answer 1 · 2016-05-27

1

Entering edit mode

Gordon Smyth 53k

@gordon-smyth

Last seen 1 hour ago

WEHI, Melbourne, Australia

Of the two packages that you have tagged, limma is the one that is designed for microarray data. I don't know what TCGA data you are using, but it is quite likely that limma can analyze the log2 values directly as a matrix. limma is designed to analyze log2-expression values.

ADD COMMENT • link 9.5 years ago Gordon Smyth 53k

0

Entering edit mode

Thank you very much. I am using microarray data of TCGA.

If I can ask one more question:

I used limma for .CEL but haven't used a matrix of log2 lowess normalised counts. Can you tell me what function does call for DEGs? will the following code be a good move?

> L_FIT <- lmFit(TheMatrix, design)

> L_FIT2 <- contrasts.fit(fit, contrast.matrix)

> L_FIT2 <- eBayes(L_FIT2)

Then using "topTable()" to get the result ?

Thanks

ADD REPLY • link 9.5 years ago Michael • 0