Dear colleagues I am doing microarray meta-analysis and I am using the E-MTAB-2325 data from the array express database. It is a single channel Agilent Whole Human Genome Microarray 4x44K 014850 G4112F. I downloaded the rawdata which is only one text file conaining Green.DerivedSignal values. How should I start the processing of this data? should I start with log2 then normalization. I have searched papers that used this data and some papers did background correction, is this possible with these data?
head(raw_data)
1 09-13-07.gpr 12bis 09-13-07.gpr 14 09-13-07.gpr
228.00000 850.00024 191.99998
30.99999 45.00000 11.00000
1184.00010 1574.99950 461.00012
1211.99960 3452.99950 1256.99990
199.99998 934.00020 152.00002
36.00000 99.99999 30.00001
•
written
by
firstorthopedicdoctor •
1 answer
We may normally read these files via limma, as follows:
data <- read.maimages(files = files, columns = 'Green.DerivedSignal')
where files is a character vector of individual sample files, and data would then be an EListRaw object, requiring further normalisation.
In this case, the authors seem to have placed all sample data into the same file, which is quite frustrating. You may have to 'hack' an EListRaw object and then attempt to normalise it by following my code here: How to process (seems) Agilent microarrry data?
Otherwise, just contact the authors and request the original raw gpr files, and / or the normalised expression matrix.
Kevin
Log in to answer this question.