MetaSKAT ver 0.40: please update the package

[Seunggeun (Shawn) Lee]

Dear all,

Updated MetaSKAT package (ver 0.40) is available on CRAN. As I posted in the previous announcement, there was a bug in the previous version of the package, which could produce a wrong p-value if you use plink files to run Meta-Burden or Meta-SKAT-O. So please update the package. 

Thanks,

Shawn

[Ken S. Lo]

Hi Shawn,
Thanks for your work on SKAT and MetaSKAT!

I just downloaded MetaSKAT ver 0.4 and the PLINK example files from one of your previous posts.  I run it and I got all NA:

"SetID" "p.value"
"1" NA
"2" NA
"3" NA
"4" NA
"5" NA
"6" NA
"7" NA
"8" NA
"9" NA
"10" NA

Here are the commands I used:

> library(MetaSKAT)
> File.Mat.vec<-rep("",3)
> File.Info.vec<-rep("",3)
> File.Mat.vec[1]<-"./01.MSSD"
> File.Info.vec[1]<-"./01.MInfo"
> File.Mat.vec[2]<-"./02.MSSD"
> File.Info.vec[2]<-"./02.MInfo"
> File.Mat.vec[3]<-"./03.MSSD"
> File.Info.vec[3]<-"./03.MInfo"
> Cohort.Info<-Open_MSSD_File_2Read(File.Mat.vec, File.Info.vec)
Number of cohorts =  3
997 samples, 10 sets, 528 SNPs and 465 unique SNPs
997 samples, 10 sets, 506 SNPs and 457 unique SNPs
997 samples, 10 sets, 521 SNPs and 467 unique SNPs
> out<-MetaSKAT_MSSD_ALL(Cohort.Info, method="optimal.adj")

Thanks
Ken

[Tati Zayats]

Hi Ken,

I've had a similar problem.

If I'm not mistaken, MetaSKAT does not need any specification of the method.

So, if you re-run your last command as

 out<-MetaSKAT_MSSD_ALL(Cohort.Info),

you will get your p-values.

I hope this helps.

[Ken S. Lo]

Thanks for your reply!

Well, I wanted to run the "optimal" version of MetaSKAT.

We finally found the problem:

out<-MetaSKAT_MSSD_ALL(Cohort.Info, method="optimal.adj")

should be

out<-MetaSKAT_MSSD_ALL(Cohort.Info, method="optimal")

[Adam Malinowski]

Why this dont work? http://wypozyczalnia-wroclaw.pl ???