Dear users,

I am trying to do a micro-array analysis using CEL files for S. aures using R.

• I have used the following libraries:

source("https://bioconductor.org/biocLite.R")
library("affy")
library("limma")
library("genefilter")

• Then I loaded the pd.s.aureus data set using the above and included it using the 

library("pd.s.aureus")

• Then using the RMA method I imported the data to eset variable. After that I tried to do a non-specific filtering using nsFilter

esetF <- nsFilter(eset, remove.dupEntrez = FALSE,
                  var.cutoff = 0.5) 

Once I do that it gives me the error 

Error: getAnnMap: package saureus not available

But I have loaded the  "pd.s.aureus" library as there is no annotation data set in the name of saureus. (https://bioconductor.org/packages/3.7/data/annotation/) 

Is there anything I am missing here ? Is there a way around this? 

Thank you very much and any suggestion is much appreciated

Neranjan

The pd.s.aureus package isn't what you need. The annotate package is looking for a package called saureus.db that it wants to use for nsFilter. I don't remember if we ever supplied that package, but we don't do so now, so you are probably better off using nsFilter without filtering on Entrez Gene IDs. You can download the annotation package from ThermoFisher and parse that to get the mappings of probeset ID to Entrez Gene ID and then just filter your ExpressionSet by hand, based on whether or not there is an Entrez Gene ID for a given probeset.

Since you have mentioned the limma package, I will add that doing variance filtering with nsFilter (by var.cutoff or otherwise) will interfere with your downstream limma analysis. You don't necessarily need to filter at all but, if you do, filtering on mean log-intensity is simpler and better in this context.

This my first time analyzing Micro array data, all comments are valuable to me. If you know any tutorial or pipeline that I should be following / or have more information please let me know as well.

Thanks

The microarray I am analyzing was done in 2010 and used Affymetrix microarray version 2.0. According to the files, I downloaded the version 31 from the Affymetrix data base, a CSV file using the following link. (https://www.affymetrix.com/support/technical/byproduct.affx?product=saureus     , file name = S_aureus.na31.annot.csv )

Since it's not a .db file version, I am not sure how to proceed in annotating the probs. 

I have 6 samples alltogether , and 3 samples in each condition. And this is what i did so far to get the p-values;

library(oligo)
​rawData <-  read.celfiles(celfiles)
#RMA normalization
normData <- rma(rawData)

library(limma)
normData2 <- normData[, normData$group %in% c("Control", "Salt")]
# Creating a design matrix
design <- model.matrix(~ normData2$group)
fit <- lmFit(normData2, design)
fit <- eBayes(fit)

And I got the pvalues and the padj values But I am unable to annotate the genes?

Is it possible to point out how the annotate the genes of S. aures ? 

Thank you very much.

Do I need to remove the spikes (= hybridsation control probes) before doing the analysis ?