Run rye fmt formatting

meyer-lab · Sep 18, 2024 · e40fb37 · e40fb37
1 parent 15b7d76
commit e40fb37
Show file tree

Hide file tree

Showing 19 changed files with 365 additions and 337 deletions.
diff --git a/pf2/data_import.py b/pf2/data_import.py
@@ -66,9 +66,9 @@ def import_data(small=False) -> anndata.AnnData:
     _, data.obs["condition_unique_idxs"] = np.unique(
         data.obs_vector("sample_id"), return_inverse=True
     )
-    data.obs["condition_unique_idxs"] = data.obs[
-        "condition_unique_idxs"
-    ].astype("category")
+    data.obs["condition_unique_idxs"] = data.obs["condition_unique_idxs"].astype(
+        "category"
+    )
 
     return data
 

diff --git a/pf2/figures/commonFuncs/plotFactors.py b/pf2/figures/commonFuncs/plotFactors.py
@@ -21,7 +21,7 @@ def plot_condition_factors(
     pd.set_option("display.max_rows", None)
     yt = pd.Series(np.unique(data.obs[cond]))
     X = np.array(data.uns["Pf2_A"])
-    
+
     XX = X
     X -= np.median(XX, axis=0)
     X /= np.std(XX, axis=0)

diff --git a/pf2/figures/commonFuncs/plotGeneral.py b/pf2/figures/commonFuncs/plotGeneral.py
@@ -11,7 +11,7 @@ def plot_avegene_per_status(
     condition="sample_id",
     cellType="cell_type",
     status1="binary_outcome",
-    status2="patient_category"
+    status2="patient_category",
 ):
     """Plots average gene expression across cell types for a category of drugs"""
     genesV = X[:, gene]
@@ -21,7 +21,7 @@ def plot_avegene_per_status(
     dataDF[status2] = genesV.obs[status2].values
     dataDF["Condition"] = genesV.obs[condition].values
     dataDF["Cell Type"] = genesV.obs[cellType].values
-    
+
     df = bal_combine_bo_covid(dataDF, status1, status2)
 
     df = pd.melt(
@@ -30,7 +30,6 @@ def plot_avegene_per_status(
 
     df = df.groupby(["Status", "Cell Type", "Gene", "Condition"], observed=False).mean()
     df = df.rename(columns={"Value": "Average Gene Expression"}).reset_index()
-
 
     sns.boxplot(
         data=df.loc[df["Gene"] == gene],
@@ -44,15 +43,23 @@ def plot_avegene_per_status(
 
     return df
 
-def bal_combine_bo_covid(df, status1: str = "binary_outcome", status2: str = "patient_category"):
+
+def bal_combine_bo_covid(
+    df, status1: str = "binary_outcome", status2: str = "patient_category"
+):
     """Combines binary outcome and covid status columns"""
-    df = df.replace({status1: {0: "L-", 
-                                1: "D-"}})
+    df = df.replace({status1: {0: "L-", 1: "D-"}})
 
-    df = df.replace({status2: {"COVID-19": "C19",
-                                "Non-Pneumonia Control": "nC19", 
-                                "Other Pneumonia": "nC19",
-                                "Other Viral Pneumonia": "nC19"}})
+    df = df.replace(
+        {
+            status2: {
+                "COVID-19": "C19",
+                "Non-Pneumonia Control": "nC19",
+                "Other Pneumonia": "nC19",
+                "Other Viral Pneumonia": "nC19",
+            }
+        }
+    )
     df["Status"] = df[status1] + df[status2]
 
     return df

diff --git a/pf2/figures/commonFuncs/plotPaCMAP.py b/pf2/figures/commonFuncs/plotPaCMAP.py
@@ -120,7 +120,7 @@ def plot_labels_pacmap(
     ds_show(result, ax)
     ax.legend(handles=legend_elements)
     ax = assign_labels(ax)
-    
+
 
 def plot_pair_wp_pacmap(
     X: anndata.AnnData,
@@ -130,16 +130,18 @@ def plot_pair_wp_pacmap(
 ):
     """Scatterplot of UMAP visualization weighted by condition or cell type"""
 
-
     points = np.concatenate(
-        ([X.obsm["weighted_projections"][:, cmp1 - 1]], [X.obsm["weighted_projections"][:, cmp2 - 1]])
+        (
+            [X.obsm["weighted_projections"][:, cmp1 - 1]],
+            [X.obsm["weighted_projections"][:, cmp2 - 1]],
+        )
     ).transpose()
 
     canvas = _get_canvas(points)
     data = pd.DataFrame(points, columns=("x", "y"))
 
     aggregation = canvas.points(data, "x", "y")
-    
+
     result = tf.shade(
         aggregation,
         how="eq_hist",
@@ -150,7 +152,6 @@ def plot_pair_wp_pacmap(
     # ax = assign_labels(ax)
 
 
-
 def plot_wp_per_celltype(
     X: anndata.AnnData, cmp: int, ax: Axes, outliers: bool = False, cellType="cell_type"
 ):
@@ -168,9 +169,7 @@ def plot_wp_per_celltype(
         ax=ax,
     )
     maxvalue = np.max(np.abs(ax.get_xticks()))
-    ax.set(
-        xticks=np.linspace(-maxvalue, maxvalue, num=5), xlabel="WP Weight"
-    )
+    ax.set(xticks=np.linspace(-maxvalue, maxvalue, num=5), xlabel="WP Weight")
     ax.set_title(cmpName)
 
 

diff --git a/pf2/figures/figure2.py b/pf2/figures/figure2.py
@@ -12,49 +12,49 @@
 
 
 def makeFigure():
-#     meta = import_meta()
-#     data = import_data()
-#     conversions = convert_to_patients(data)
+    #     meta = import_meta()
+    #     data = import_data()
+    #     conversions = convert_to_patients(data)
 
     axs, fig = getSetup((6, 6), (2, 1))
 
-#     ranks = np.arange(5, 65, 5)
-#     r2xs = pd.Series(0, dtype=float, index=ranks)
-#     accuracies = pd.Series(0, dtype=float, index=ranks)
-#     for rank in ranks:
-#         fac, r2x = pf2(data, rank, do_embedding=False)
-#         patient_factor = pd.DataFrame(
-#             fac.uns["Pf2_A"],
-#             index=conversions,
-#             columns=np.arange(fac.uns["Pf2_A"].shape[1]) + 1,
-#         )
-#         if meta.shape[0] != patient_factor.shape[0]:
-#             meta = meta.loc[patient_factor.index, :]
+    #     ranks = np.arange(5, 65, 5)
+    #     r2xs = pd.Series(0, dtype=float, index=ranks)
+    #     accuracies = pd.Series(0, dtype=float, index=ranks)
+    #     for rank in ranks:
+    #         fac, r2x = pf2(data, rank, do_embedding=False)
+    #         patient_factor = pd.DataFrame(
+    #             fac.uns["Pf2_A"],
+    #             index=conversions,
+    #             columns=np.arange(fac.uns["Pf2_A"].shape[1]) + 1,
+    #         )
+    #         if meta.shape[0] != patient_factor.shape[0]:
+    #             meta = meta.loc[patient_factor.index, :]
 
-#         acc, _ = predict_mortality(patient_factor, meta)
-#         r2xs.loc[rank] = r2x
-#         accuracies.loc[rank] = acc
-#         r2xs.to_csv("/home/jchin/BAL-Pf2/output/r2x_v_rank_no_ig.csv")
-#         accuracies.to_csv("/home/jchin/BAL-Pf2/output/acc_v_rank_no_ig.csv")
+    #         acc, _ = predict_mortality(patient_factor, meta)
+    #         r2xs.loc[rank] = r2x
+    #         accuracies.loc[rank] = acc
+    #         r2xs.to_csv("/home/jchin/BAL-Pf2/output/r2x_v_rank_no_ig.csv")
+    #         accuracies.to_csv("/home/jchin/BAL-Pf2/output/acc_v_rank_no_ig.csv")
 
-#         gc.collect()
+    #         gc.collect()
 
-#     # R2X Plots
+    #     # R2X Plots
 
-#     axs[0].plot(r2xs.index.astype(int), r2xs)
-#     axs[0].set_xticks(r2xs.index.astype(int))
-#     axs[0].grid(True)
+    #     axs[0].plot(r2xs.index.astype(int), r2xs)
+    #     axs[0].set_xticks(r2xs.index.astype(int))
+    #     axs[0].grid(True)
 
-#     axs[0].set_ylabel("R2X")
-#     axs[0].set_xlabel("Rank")
+    #     axs[0].set_ylabel("R2X")
+    #     axs[0].set_xlabel("Rank")
 
-#     # Accuracy Plots
+    #     # Accuracy Plots
 
-#     axs[1].plot(accuracies.index, accuracies)
-#     axs[1].set_xticks(accuracies.index.astype(int))
-#     axs[1].grid(True)
+    #     axs[1].plot(accuracies.index, accuracies)
+    #     axs[1].set_xticks(accuracies.index.astype(int))
+    #     axs[1].grid(True)
 
-#     axs[1].set_ylabel("Accuracy")
-#     axs[1].set_xlabel("Rank")
+    #     axs[1].set_ylabel("Accuracy")
+    #     axs[1].set_xlabel("Rank")
 
     return fig
diff --git a/pf2/figures/figure4.py b/pf2/figures/figure4.py
@@ -35,10 +35,8 @@ def makeFigure():
         boot_meta = meta.iloc[boot_index, :]
         _, (covid_plsr, nc_plsr) = predict_mortality(boot_factor, boot_meta)
 
-        covid_coefficients.loc[trial + 1, covid_plsr.coef_.index] = \
-            covid_plsr.coef_
-        nc_coefficients.loc[trial + 1, nc_plsr.coef_.index] = \
-            nc_plsr.coef_
+        covid_coefficients.loc[trial + 1, covid_plsr.coef_.index] = covid_plsr.coef_
+        nc_coefficients.loc[trial + 1, nc_plsr.coef_.index] = nc_plsr.coef_
 
     axs, fig = getSetup((8, 4), (1, 1))
     ax = axs[0]

diff --git a/pf2/figures/figureA1.py b/pf2/figures/figureA1.py
@@ -1,4 +1,4 @@
-"""Figure A1: Condition, eigen-state, and gene factors, 
+"""Figure A1: Condition, eigen-state, and gene factors,
 along with PaCMAP labeled by cell type"""
 
 import anndata
@@ -23,12 +23,14 @@ def makeFigure():
     X.uns["Pf2_A"] = correct_conditions(X)
     add_obs(X, "patient_category")
     add_obs(X, "binary_outcome")
-    plot_condition_factors(X, ax[0], cond="sample_id", cond_group_labels=pd.Series(label_all_samples(X)))
+    plot_condition_factors(
+        X, ax[0], cond="sample_id", cond_group_labels=pd.Series(label_all_samples(X))
+    )
     ax[0].yaxis.set_ticklabels([])
     plot_eigenstate_factors(X, ax[1])
     plot_gene_factors(X, ax[2])
     ax[2].yaxis.set_ticklabels([])
- 
+
     df = X.obs[["patient_category", "binary_outcome"]].reset_index(drop=True)
     df = bal_combine_bo_covid(df)
     X.obs["Status"] = df["Status"].to_numpy()
@@ -50,24 +52,19 @@ def label_all_samples(X: anndata.AnnData):
     for i, sample in enumerate(pd.unique(X.obs["sample_id"])):
         bo = pd.unique(X[X.obs.sample_id.isin([sample])].obs["binary_outcome"])
         if bo == 0:
-            bo = "L-" 
+            bo = "L-"
         else:
             bo = "D-"
         bo_only[i] = bo
 
-        pc = pd.unique(X[X.obs.sample_id.isin([sample])].obs["patient_category"]) 
+        pc = pd.unique(X[X.obs.sample_id.isin([sample])].obs["patient_category"])
         if pc == "COVID-19":
             pc = "C19"
-        else: 
+        else:
             pc = "nC19"
         pc_only[i] = pc
-
-    for i in range(len(labels_samples)):
-        labels_samples[i] = bo_only[i]+pc_only[i]
-
-
-    return labels_samples
-
 
+    for i in range(len(labels_samples)):
+        labels_samples[i] = bo_only[i] + pc_only[i]
 
-
+    return labels_samples