[BioC] Error in kooperberg procedure

[Matthew Ritchie]

Dear Mick,

Yes, a 0 value for the background intensity will cause this error.  I 
have modified the code in the latest version of limma (1.7.7) to handle 
0's, so hopefully this is no longer a problem.  Best wishes,

Matt Ritchie

michael watson (IAH-C) wrote:

>Hello
>
>I ran this code and got an error:
>
>>genepixFiles <- targets$Name
>>read.series(genepixFiles, suffix=NULL,skip=26, sep="\t")
>>layout <- RG$printer
>>RGmodel <- kooperberg(genepixFiles, layout=layout) 
>>
>Error in integrate(.numeratorBayesianAdjustedFG, ifelse((fg - bg - 4 *
>: 
>        a limit is missing
>
>I guess this may be related to my earlier problem, where I have BG
>estimates = zero for one of my arrays, but this is a little trickier to
>solve as the data is in data.frames, and I cannot so easily add a small
>value to Rb and Gb....
>
>Mick
>

	[[alternative HTML version deleted]]