Save Andrew's tutorial

pf2rnaseq/figures/figureLupus1.py

@@ -2,7 +2,7 @@
 XX:
 """
 from matplotlib.pylab import eig
-import plotly.express as px
+# import plotly.express as px
 from anndata import read_h5ad
 from ..imports import pseudobulk_lupus
 from .common import subplotLabel, getSetup
@@ -14,30 +14,51 @@
 import matplotlib.pyplot as plt
 
 
-
-
-
-
 def makeFigure():
     """Get a list of the axis objects and create a figure."""
     # Get list of axis objects
-    ax, f = getSetup((30, 8), (2, 4))
+    ax, f = getSetup((9, 9), (3, 3))
 
     # Add subplot labels
     subplotLabel(ax)
 
 
     X = read_h5ad("/opt/andrew/lupus/lupus_fitted_ann.h5ad")
+
     bulk_matrix, bulk_tensor = pseudobulk_lupus(X)
-    print(bulk_matrix )
-    # bulk_matrix_genes_only = bulk_matrix.iloc[:, 2:-1]
-
-    # pca = PCA(n_components=2)
-    # principalComponents = pca.fit_transform(bulk_matrix_genes_only)
-    # print(np.shape(principalComponents))
-
-    # sns.scatterplot(data=data, x="PC1", y="PC2", hue="Status", ax=ax[0])
-
+
+
+    bulk_matrix = bulk_matrix.iloc[[0, 1, 10, 11], :]
+    bulk_matrix = bulk_matrix.iloc[:, [0, 1, 2, 3, 4, -1]]
+
+    bulk_matrix_genes_only = bulk_matrix.iloc[:, 2:5]
+    print(bulk_matrix_genes_only)
+
+    pca = PCA(n_components=2)
+    scores = pca.fit(bulk_matrix_genes_only).transform(bulk_matrix_genes_only)
+    loadings = pca.components_.T
+
+    print(np.shape(scores))
+    print(np.shape(loadings))
+
+    scores_df = pd.DataFrame(data = scores, columns = ["PC1", "PC2"])
+    loadings_df = pd.DataFrame(data = loadings, columns = ["PC1", "PC2"])
+
+    loadings_df["Gene"] = bulk_matrix_genes_only.columns.to_numpy()
+    sns.scatterplot(data=loadings_df, x="PC1", y="PC2", hue="Gene", ax=ax[4])
+
+    print(loadings_df)
+
+
+    # sns.scatterplot(data=scores_df, x="PC1", y="PC2", ax=ax[0])
+
+    # scores_df["Status"] = bulk_matrix["Status"].to_numpy()
+    # scores_df["Condition"] = bulk_matrix["Condition"].to_numpy()
+    # scores_df["Cell Type"] = bulk_matrix["Cell Type"].to_numpy()
+
+    # sns.scatterplot(data=scores_df, x="PC1", y="PC2", hue="Status", ax=ax[1])
+    # sns.scatterplot(data=scores_df, x="PC1", y="PC2", hue="Condition", ax=ax[2])
+    # sns.scatterplot(data=scores_df, x="PC1", y="PC2", hue="Cell Type", ax=ax[3])
 
     # url = (bulk_matrix)