how to calculate log average intensity for MA plot?

Question

[KABILAN]

Hello, I have a proteomics expression dataset. The head portion of the data will be like,

data <- structure(list(`Fasta headers` = c(">O76070ups|SYUG_HUMAN_UPS Gamma-synuclein (Chain 1-127) - Homo sapiens (Human)", 
">Q06830ups|PRDX1_HUMAN_UPS Peroxiredoxin 1 (Chain 2-199) - Homo sapiens (Human)", 
">P06396ups|GELS_HUMAN_UPS Gelsolin (Chain 28-782) - Homo sapiens (Human);>Q3SX14 TREMBL:Q3SX14 (Bos taurus) Similar to Gelsolin", 
">P02768-1 SWISS-PROT:P02768-1 Tax_Id=9606 Gene_Symbol=ALB Isoform 1 of Serum albumin precursor;>P02768ups|ALBU_HUMAN_UPS Serum albumin (Chain 26-609) - Homo sapiens (Human)", 
">P02741ups|CRP_HUMAN_UPS C-reactive protein (Chain 19-224) - Homo sapiens (Human)", 
">P16083ups|NQO2_HUMAN_UPS Ribosyldihydronicotinamide dehydrogenase [quinone] (Chain 2-231) - Homo sapiens (Human)"
), A1 = c(28.8484762528371, 28.5593417132562, 29.8009889375404, 
30.236308349045, 26.8634920403497, 29.2127142763584), A2 = c(28.6976154934535, 
28.5259670144823, 29.7664700243508, 30.1817239029611, 26.8135256143612, 
29.0836758932669), A3 = c(28.6907247615967, 28.4075268367718, 
29.945806961862, 30.1906689352863, 26.8775178221577, 29.1529637057232
), B1 = c(21.4759289585346, 21.8116726154379, 21.0287288705184, 
21.6755517309807, 22.3711955096869, 20.5319556862522), B2 = c(21.2438429926591, 
21.5540102900844, 21.1130287737854, 21.2063689577213, 22.5489466679164, 
20.622315219241), B3 = c(21.9123487685507, 22.2790808524578, 
21.1321045998028, 22.3747058136723, 22.3639090145369, 20.6142641267144
)), row.names = c("1", "2", "3", "4", "5", "6"), class = "data.frame")

And I want to analyse it for differential expression analysis. For that I have to plot MA plot. I got the log fold change value from the following code,

library(limma)
library(edgeR)

cts <- as.data.frame(data)

#Changing the first column as a row name
data2 <- cts [,-1]
Protein_info <- cts[,1]
rownames(data2) <- make.names(Protein_info, unique = TRUE)

#Creating the model matrix
eset <- data2
Groups <- model
design <- model.matrix(~0+factor(Groups))
colnames(design)<- c("A", "B")

#Fitting the model
fit <-lmFit(eset, design)
fit <- eBayes(fit)
options (digits = 2)
res <- topTable(fit, number= Inf, adjust.method = "none", coef = 1)

#Making the table with logFC and p-values
res1 <- as.data.frame(res)
my_data <- as_tibble(res1)
res_table2 <-my_data%>%select(1,3)
res_table2 <- as.data.frame(res_table2)
id_name <- as.data.frame(Protein_info)
res_table <- cbind(id_name, res_table2)
res_table$logFC

But I couldn't get average log intensity values [A = 1/2(logB+logA)] at base 2, for plotting in the x-axis of MA plot. I am new to this field. So, kindly give some idea for getting this values.

It may seems like a simple question. Since I am new to this area, please help me.. Thank you in advance.

Answer 1

[ATpoint]

Its the AveExpr column in the topTable output which is nothing but a rowMeans of the input data. I would guess that your data are already log2 based on the magnitude of values and the fact that it's proteomics.

[KABILAN]

Okay. Sorry I am not getting your answer. What is the column name of log average intensity values from the above code? and from which dataset?

[ATpoint]

The AveExpr column in the output of the topTable function.