diff --git a/inst/pages/clustering.qmd b/inst/pages/clustering.qmd
index b3665175..cf5fcb95 100644
--- a/inst/pages/clustering.qmd
+++ b/inst/pages/clustering.qmd
@@ -304,11 +304,13 @@ features.
 #| fig-height: 2
 
 # Get the estimates on how much each phylum contributes on each cluster
-best_model <- metadata(altExp(tse, "prevalent"))$DMM$dmm[2]
-drivers <- as.data.frame(best_model[[1]]@fit$Estimate)
+models <- metadata(altExp(tse, "prevalent"))$DMM
+n_clusters <- models[["best"]]
+best_model <- models[["dmm"]][[n_clusters]]
+drivers <- fitted(best_model, scale = TRUE)
 
 # Plot by utilizing miaViz's function plotLoadings
-plotLoadings(as.matrix(drivers), ncomponents = 2)
+plotLoadings(drivers, ncomponents = 2)
 ```
 
 As well as in hierarchical clustering, we can also visualize the clusters by