limma. After running a differential expression analysis... Can you get the individual gene expression levels between individual contrasts for each of the DE genes?
2
0
Entering edit mode
arfranco ▴ 130
@arfranco-8341
Last seen 21 hours ago
European Union

I am running a DE analysis with limma using a one-color microarray, and got the toptables indicating those genes which are differentially expressed.

My colleague (and also myself) want to know, however ,the level of gene expression between individual contrasts for each of the genes that give rise to the LogFoldChange value

We need to know whether a DE gene changes from a low, medium or high expression in the control to a lower or even a higher expression in the other condition included in the contrast

To my understanding, AveExp gives the log2 Expression mean level of each of the gene in all the compared arrays.

Is there at least a way to know the normalized level of expression of the control array in each of the genes of the array ?.

0
Entering edit mode
arfranco ▴ 130
@arfranco-8341
Last seen 21 hours ago
European Union

I'll try to answer myself. Please confirm whether I am right or not..

Let's assume that  data_normalized is a Elist obtained after substracting the background, after a normalizeBetweenArrays normalization, and after running avereps function to average the expression levels..

I believe that the information I need is included in the data_normalized$E field? Am I right? ADD COMMENT 0 Entering edit mode 1. It's hard to tell what you're doing without code, read the posting guide. 2. Looking at ?"EList-class" would probably yield the answers you're looking for. ADD REPLY 0 Entering edit mode arfranco ▴ 130 @arfranco-8341 Last seen 21 hours ago European Union I believe that this can be the solution datos_norm1 is the Elist file containing the subtracted, normalized and averaged data processed through the limma functions. I am analyzing a total of 24 arrays ​my_data <- as.data.frame(cbind(datos_norm1$genes, datos_norm1$E)) # Join gene data (names..) and expression values # my_data[row, 7:30] # row is any row consisting in a gene; 7:30 contains the expression values of a total of 24 arrays I am analyzing # which(my_data$SystematicName == "gene01176-v1.0-hybrid") # Allows me to find the row for the gene I am interested in
# rowMeans(my_data[row, 7:30]) # Gives me the AveExp value I see in the toptables

So I only need to use dplyr or any other package to  organize and include in the toptable the control and the sample expression values that are eventually used in the DE analysis. For us, this has no statistical, but biological value

0
Entering edit mode

Wondering why in that Elist file, there are two rows for each of the genes.. ¿what is the meaning of that?

Daa values are very much similar between the two rows having the same gene name.. What is the origen of that ? Where and how is these data obtained