In isglobal-brge/dsExposomeClient: DataSHIELD client site Exposome functions

A HCPC (hierarchical classification on principle components) can be performed on the exposures of the exposome dataset. To do so, there's the ds.exposome_HCPC. By default, the number of clusters discovered is the one with the higher relative loss of inertia (i(clusters n+1)/i(cluster n)), the user can also specify the number of clusters with the argument nb.clust. This function has to be passed the server object that holds the results of a PCA.

ds.exposome_pca("exposome_object", fam = c("Metals", "Noise"))
hcpc=ds.exposome_HCPC("ds.exposome_pca.Results")[[1]]

The object returned by the servers can be plotted using the FactoMineR::plot.HCPC function.

FactoMineR::plot.HCPC(hcpc)
FactoMineR::plot.HCPC(hcpc, choice = "tree")
FactoMineR::plot.HCPC(hcpc, choice = "bar")
FactoMineR::plot.HCPC(hcpc, choice = "map")