[adegenet-forum] DAPC

[Kirsty Medcalf]

Hi Mariana, and  Jombart

Thank you for your suggestions.  I was finally successful with creating and
customising  the scatter plot.  I am now attempting to cross validate the
DAPC analysis using the code:

x1 <- LDA.scores
mat <- as.matrix(x1, method="mean")
grp <- x1
xval <- xvalDapc(mat, grp2, n.pca.max = 300, training.set = 0.7,
                 result = "groupMean", center = TRUE, scale = FALSE,
                 n.pca = NULL, n.rep = 30, xval.plot = TRUE)

The output is this error:

Error in sort.list(y) : 'x' must be atomic for 'sort.list'
Have you called 'sort' on a list?

If this is possible, I have been playing around with this code for a few
days and checked online.  Would you happen to have any idea why this is
occurring.  Thank you if you can help.

Best wishes

Kirsty

Kirsty Medcalf

kirsty.m.medcalf at gmail.com

+447963374030

skype contact: kirsty.medcalf

On Wed, Sep 23, 2015 at 7:11 AM, Jombart, Thibaut <t.jombart at imperial.ac.uk>
wrote:

> Hi there,
>
> sorry I did not go into the details, but checking for potential mistakes:
>
> #1
>  'na.action' is not an argument in find.clusters
>
> #2
>
> x<-Just.V4[2:13]
>
> this looks like a vector to me, not a matrix/data.frame; I am not quite
> sure what you expect 'grp' to be then.
>
> As for producing scatterplots with varying factors, see argument 'grp' in
> ?scatter.dapc
>
> Cheers
> Thibaut
> ==============================
> Dr Thibaut Jombart
> MRC Centre for Outbreak Analysis and Modelling
> Department of Infectious Disease Epidemiology
> Imperial College - School of Public Health
> Norfolk Place, London W2 1PG, UK
> Tel. : 0044 (0)20 7594 3658
> http://sites.google.com/site/thibautjombart/
> http://sites.google.com/site/therepiproject/
> http://adegenet.r-forge.r-project.org/
> Twitter: @thibautjombart
>
>
> ------------------------------
> *From:* adegenet-forum-bounces at lists.r-forge.r-project.org [
> adegenet-forum-bounces at lists.r-forge.r-project.org] on behalf of Kirsty
> Medcalf [kirsty.m.medcalf at gmail.com]
> *Sent:* 23 September 2015 05:45
> *To:* adegenet-forum at lists.r-forge.r-project.org
> *Subject:* [adegenet-forum] DAPC
>
>
> Dear Forum
>
> This is my first post, so I would like to thank you for your patience.  The
> multivariate data that I am using contains two categorical grouping factors
> (V4 or G8) under the column family (response variable) and 12 accompanying
> predictor variables. The data is called LDA.scores and is found at the
> bottom of my Stack Overflow page by following the link below, which shows
> my attempted step-by-step logic and figures.
>
>
> http://stackoverflow.com/questions/32704902/discriminant-analysis-of-principal-components-and-how-to-graphically-show-the-di
>
> I have been attempting to graphically show the distance of data points to
> its multivariate centroid using DAPC analysis and the function `scatter' in
> the `adegenet' package in R. After splitting the two categorical factors
> into two separate data frames (coding below), I attempt to produce these
> scatterplot. I understand this package is used for the analysis of genetic
> markers, however, I am also under the impression that all types of
> multivariate data can be analysed using this package. I tried to manipulate
> the data but to no avail.
>
> Code used to produce figure
> *Split the dataframe into just V4 and G8
>
> Just.V4<-LDA.scores[LDA.scores$Family=="V4",]
> Just.G8 <-LDA.scores[LDA.scores$Family=="G8",]
>
> #Attempt to produce a scatterplot for the categorical factor V4
> library(adegenet)
> x<-Just.V4[2:13]
>
> *Find the clusters
>
> grp<-find.clusters(x, max.n.clust=12, na.action="omit")
>
> The next step is the perform the discriminant analysis of principal
> components
>
>  dapc1<-dapc(x, grp$grp)
>  scatter(dapc1)
>
> I have tried many different combinations of code and here are some of the
> error messages
>
> Error in dapc.data.frame(x, grp1$grp1) : Inconsistent length for grp
> Warning in find.clusters.data.frame(as.data.frame(x), ...) :
> NAs introduced by coercion
> Error in if (n.pca >= N) warning("number of retained PCs of PCA is  greater than N") :
> missing value where TRUE/FALSE needed
>
>
> If anyone has a solution in terms of how to produce two figures for each
> categorical factor which illustrates the clusters (12 parameters measured)
> to its multivariate centroid, then thank so much. I have followed lots of
> tutorials, searched online and read papers, and still do not understand
> these error and warning messages.
>
> Thank you if anyone can help.
>
> Best wishes,
> Kaikash
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/adegenet-forum/attachments/20150926/344bcdfa/attachment-0001.html>