Hello,
I have a question which was part answered on another post but I have follow up question. Dichotomizing Gene Expression Data Based On The Median
I am using for loop and most use as doing calculation further on which require it.
my question is
since I am using for loop, I need to access the gene expression value for each gene separately to do the comparison to median.
geneA 0.54744
gene B 0.5596
and so on....
I have the normalised data but that's a matrix of the genes and the samples what each got. for example,
genes sample 1 sample 2 sample 3 sample 4
7945460 7.471390 7.256158 7.287770 7.545794
7945462 7.609366 7.528324 7.324294 7.310791
7945475 5.375443 5.749566 5.519073 5.806861
How / where is the final number for the gene? so I can calculate comparison.
1 answer
It's not clear what you are asking, but you can greatly simplify your calculations:
#make some fake data. you call this current_study
# 100 genes * 10 samples.
expr = matrix(rnorm(1000), nrow=100, ncol=10)
rownames(expr) = paste0("gene", 1:100)
colnames(expr) = paste0("sample", 1:10)
# get the median for each gene.
meds = apply(expr, 1, median)
# you called this di_matrix
gtmed = expr > meds
where gtmed now contains TRUE if that entry is above the median for that row.
Just use what brentp posted and then take the median of medians. What he posted will give you one value per gene/probe/whatever. I should note that I'm not a fine of dichotomous expressed/unexpressed methods (the mas5 method is OK).
•
written
by
Devon Ryan
Log in to answer this question.