biotyper: Biotyping your commununity

In tapj/biotyper: BiotypeR: an R package to biotype a community

Description

This function permit to cluster your sites/samples (columns) in k biotypes according distribution of features (species or functions in rows), calculate cluster and clasification index, provide multivariate analysis as PCA, between class and dPCoA, and plot everything. The same function take in charge ecosimulation object.

Usage

biotyper(x, ...)

## S3 method for class 'data.frame'
biotyper(obs, k=3 , distance.jsd=NULL, cluster=NULL, manalysis=FALSE,  plot=FALSE, nf=3, no.unassigned=TRUE,...)
## S3 method for class 'ecosimulation'
biotyper(obs, k , ...)
## S3 method for class 'biotyper'
plot(biotyper.obj, xax=1, yax=2, potatoes=FALSE, ...)
## S3 method for class 'biotyper.obj.simul'
print(biotyper.obj.simul)
## S3 method for class 'biotyper.obj.simul'
summary(biotyper.obj.simul)

Arguments

obs	A data.frame with sites as columns and species/function as rows.
k	number of clusters
distance.jsd	a distance matrix of sites.if NULL a jsd distance matrix will be created using dist.JSD
cluster	a numeric or character vector to force your cluster
manalysis	logical. if FALSE, no multivariate analysis are done
plot	logical. if FALSE, nothing is plotted
nf	Numbers of component, if you do not know keep nf=3 by default.
no.unassigned	logical. Do you have an unassigned features (on the first row) ? If TRUE, multivariate analysis will remove it during PCA analysis.
...	not used
biotyper.obj	an object of class "biotyper.obj", i.e. a biotyper object
biotyper.obj.simul	an object of class "biotyper.obj.simul", i.e. a biotyper object
potatoes	logical. if TRUE, s.potatoes will be used to represent biotypes
xax	the component number for the x axis
yax	the component number for the y axis

Details

Have a look to Arumugam et al.Nature.2011.

Value

Acoording the class of object (i.e a data.frame or ecosimulation), an object of class biotyper.obj or biotyper.obj.simul, basically including a list including elements

biotypes	input or generated biotypes using pamk function from codefpc
obs.silhouette	silhouette coefficient
PCA	Principal component analysis using ade4
BET	Between class analysis using ade4
dPCoA	Double principle coordinate analysis using ade4

Author(s)

Julien Tap

References

Arumugam et al. Enterotypes of of the human gut microbiome. Nature 2011. doi:10.1038/nature09944

See Also

objects to See Also as test.biotypes , CH.index , ecosimulator

Examples

data(Titanium16S)
Titanium16S.jsd=dist.JSD(Titanium16S)
ch=CH.index(Titanium16S,Titanium16S.jsd,k=1:10)
k=which(ch==max(ch))
biotyper(Titanium16S, k, distance.jsd=Titanium16S.jsd, manalysis=TRUE, plot=TRUE)


## Not run: 
## Not run:
data(Titanium16S)
Titanium.jsd=dist.JSD(Titanium16S)
ch=CH.index(Titanium16S,Titanium16S.jsd,k=1:10)
k=which(ch==max(ch))
ecosimulation=ecosimulator(Titanium16S, 100)
biotyper.obj.simul=biotyper(ecosimulation, k)
summary(biotyper.obj.simul)
## End (Not run)

## End(Not run)

tapj/biotyper documentation built on June 4, 2018, 7:33 p.m.