Kmeans: assign cluster to each variate

[ciropom]

Hello, I've read

https://github.com/sib-swiss/dsSwissKnife-example/blob/main/dsSwissKnife_example.R

And I successfully performed PCA and KNN. Now I want to plot the first two PCA components using ggplot, coloring each variates with a colour indicating the cluster he belongs to.

When I compute a kmeans using plain R, I can use '$cluster' to map a sample to a cluster, but with dss version the cluster attribute is different

# plain R kmeans
> fit.km$cluster
  [1] 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 1 1 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2
 [61] 2 2 2 2 2 2 2 2 2 2 1 2 2 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
[121] 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

# dssKmeans
> knn$global$cluster
    1     2 
13072  8335 

How can I assign cluster label to my samples?

I tried also to layer the kmeans cluster classification onto the PCA biplots, as shown in the example but I've found the result confusing and difficult to interpret.

Thank you Danilo

[ciropom]

Hello, just bumping up this discussion. Is there any way to get per-sample cluster labels from dssKmeans?

Thank you

[IulianD]

Hello, sorry, I somehow missed this with the holidays an others. I'll look into it and get back to you.

[ciropom]

thank you!