Try UMAP

app.py

@@ -23,6 +23,7 @@ import matplotlib.colors as mcolors
 import zipfile
 import tempfile
 from kneed import KneeLocator
+import umap
 
 
 class RelativeScaler(BaseEstimator, TransformerMixin):
@@ -561,6 +562,12 @@ def compute_global_regression(df_combined, embedding_cols, tsne_params, df_f1, r
             ("pca", PCA(n_components=N_COMPONENTS)),
             ("rel_scaler", RelativeScaler())
         ])
+    elif reduction_method == "UMAP":
+        reducer = umap.UMAP(n_components=N_COMPONENTS, 
+                       random_state=42, 
+                       n_neighbors=15, 
+                       min_dist=0.1,
+                       metric='cosine')
     else:
         reducer = TSNE(n_components=2, random_state=42, 
                          perplexity=tsne_params["perplexity"], 
@@ -832,7 +839,7 @@ def run_model(model_name):
         return
 
     st.markdown('<h6 class="sub-title">Select Dimensionality Reduction Method</h6>', unsafe_allow_html=True)
-    reduction_method = st.selectbox("", options=["PCA", "t-SNE"], key=f"reduction_{model_name}")
+    reduction_method = st.selectbox("", options=["PCA", "t-SNE", "UMAP"], key=f"reduction_{model_name}")
     
     distance_metric = st.selectbox("Select Distance Metric:", 
                                    options=["Euclidean", "Wasserstein", "KL"],