Entering edit mode
hi,
i am normalizing my CEL files with mas5 but i had negative values then i did like below befor log transformation but
library(affy)
Data<-ReadAffy()
eset<-mas5(Data)
norm.data <- sweep(exprs(eset), abs(min(exprs(eset))) + 5, "+")
Error in if (lstats > prod(dimmargin)) { :
missing value where TRUE/FALSE needed
then what to do?? tell me a solution please
thank you
i could not get anything :(
what i did wrong please?
It's likely regarding your second argument to the call to
sweep, but if all you are trying to do is to add the same small constant to all the values in your matrix, you could also simply do:thank you, I was going to do log2 transformation on my norm.data because mas5 result were too big while i had negative values then by your tip i don't have negative value anymore but after log transformation i have negative value again
> mycounts <- log2(norm.data)
> head(mycounts[,1:4])
Col-0 24h primed.CEL.CEL Col-0 24h unprimed.CEL.CEL
244901_at 9.098992 7.7423665
244902_at 6.509487 5.3400972
244903_at 9.173553 9.2231098
244904_at -1.496849 -0.1344298
244905_at 1.169922 0.7548389
244906_at 7.473863 5.3282473
Col-0 48h primed.CEL.CEL Col-0 48h unprimed.CEL.CEL
244901_at 7.936468 7.5260384
244902_at 5.147705 5.4040515
244903_at 7.568202 7.8444385
244904_at -1.163557 -0.7156607
244905_at 1.657202 1.9424205
244906_at 6.033295 5.3367284
> what is the solution please
thank you
There is no solution, as none is needed. Taking logs of a number between 0 and 1 will result in a negative value. But this isn't a problem for any down-stream processing steps.
thank you very much