[BioC] missing value where TRUE/FALSE needed (metaArray)

[mjonczyk]

Dear Hyungwon, dear All,

Data is on log scale.
I've checked other two studies which I'm going to merge to metaanalysis
and variances' ranges are similar.

study 2
Min.        1st Qu.       Median       Mean      3rd Qu.       Max.
0.002281   0.106800      0.309800    0.783000   0.896700      19.610000

study 3
Min.        1st Qu.       Median       Mean      3rd Qu.       Max.
0.000472   0.047810      0.120300     0.344400  0.307800      20.290000

So what is minimal variance which don't cause the error?

Thank You,
Best Wishes


---
Maciej Jonczyk,
Department of Plant Molecular Ecophysiology
Institute of Plant Experimental Biology
Faculty of Biology, University of Warsaw
02-096 Warsaw, Miecznikowa 1

On Mon, 26 Sep 2011 15:02:23 +0800, Hyungwon Choi wrote:
> Also, if your data is not on log scale, please take logs. The model
> is under certain uninformative prior and if the data goes out of its
> range, the sampler may diverge.
>
> Please don't hesitate to let me know if it still doesn't work out for 
> you.
>
> Hyungwon
>
>
> On 09/26/2011 02:59 PM, mjonczyk wrote:
>> Thank You All for fast reply.
>>
>> I've checked row variance and indeed some values must be close to 0.
>>> xv=rowVars(exprs(AFY2me))
>>> summary(xv)
>>       Min.    1st Qu.   Median      Mean       3rd Qu.   Max
>> 0.001699   0.044130   0.105300   0.310600   0.258800   20.960000
>>
>> Actually this low variance looks strange for me - it is data from
>> experiment with two treatments
>> fungal inoculation and mock inoculation. So I'm suprised - I should
>> check everything once more.
>>
>> I'll also check the code provided by Reema - thank you!
>>