[R] bootstrapping problem

[Uwe Ligges]

On 29.04.2011 09:01, mimato wrote:
> I want to classify bipolar neurons in human cochleas and have data of the
> following structure:
>
> Vol_Nuc Vol_Soma
> 1   186.23   731.96
> 2   204.58  4370.96
> 3   539.98  7344.86
> 4   477.71  6939.28
> 5   421.22  5588.53
> 6   276.61  1017.05
> 7   392.28  6392.32
> 8   424.43  6190.13
> 9   256.41  3850.51
> 10  249.17  3118.14
> 11  276.97  3037.29
> 12  295.30  3703.76
> 13  314.43  5265.97
> 14  301.15  5781.73
>
> I already worked with Matlab (I´m not a programmer) and created nice
> colourcoded dendrograms and also made some verifications of them. I started
> now working with R and bootstrapped data with a library named pvclust. It
> worked and R computed ...
>
> here is the code:
>
> library(pvclust)
>
> data =
> data.frame(Vol_Nuk=c(186.23,204.58,539.98,477.71,421.22,276.61,392.28,424.43,256.41,249.17,276.97,295.3,314.43,301.15),
> Vol_Soma=c(731.96,4370.96,7344.86,6939.28,5588.53,1017.05,6392.32,6190.13,3850.51,3118.14,3037.29,3703.76,5265.97,5781.73))
>
> plot(data)
> result<-pvclust(data,nboot=100)
> plot(result)
>
> It is also not working using following commands:
>
> cluster.bootstrap<- pvclust(Raw, nboot=1000, method.dist="abscor")
> plot(cluster.bootstrap)
> pvrect(cluster.bootstrap)
>
> I always get the following problem:
>
> mistake in plot.hclust(x$hclust, main = main, sub = sub, xlab = xlab, col =
> col,  :
> invalid input for Dendrogram
>
> Does anyone has an idea whats wrong...


Yes: You have only 2 variables and when clustering these 2 variables it 
makes no sense to plot this. My I guess you want to cluster observations 
rather than variables? Then transpose your data before applying pvclust.

Best,
Uwe Ligges



> Thanks a lot!!
>
> --
> View this message in context: http://r.789695.n4.nabble.com/bootstrapping-problem-tp3483068p3483068.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.