CRC analysis

pf2rnaseq/figures/common.py

@@ -19,7 +19,7 @@
 matplotlib.rcParams["ytick.major.pad"] = 1.0
 matplotlib.rcParams["xtick.minor.pad"] = 0.9
 matplotlib.rcParams["ytick.minor.pad"] = 0.9
-matplotlib.rcParams["legend.handletextpad"] = 0.5
+matplotlib.rcParams["legend.handletextpad"] = 0.4
 matplotlib.rcParams["legend.handlelength"] = 0.5
 matplotlib.rcParams["legend.framealpha"] = 0.5
 matplotlib.rcParams["legend.markerscale"] = 0.7

pf2rnaseq/figures/commonFuncs/plotFactors.py

@@ -15,18 +15,15 @@ def plot_condition_factors(
     data: AnnData,
     ax: Axes,
     cond_group_labels: Optional[pd.Series] = None,
-    ThomsonNorm=False,
+    condition_label="Condition",
     groupConditions=False,
 ):
     """Plots Pf2 condition factors"""
     pd.set_option("display.max_rows", None)
-    yt = pd.Series(np.unique(data.obs["Condition"]))
+    yt = pd.Series(np.unique(data.obs[condition_label]))
     X = np.array(data.uns["Pf2_A"])
 
     X = np.log10(X)
-    if ThomsonNorm is True:
-        controls = yt.str.contains("CTRL")
-        X = X[controls]
 
     X -= np.median(X, axis=0)
     X /= np.std(X, axis=0)
@@ -46,14 +43,14 @@ def plot_condition_factors(
         # extra padding to leave room for the row colors
         # get list of colors for each label:
         colors = sns.color_palette(
-            n_colors=pd.Series(cond_group_labels).nunique()
+            n_colors=cond_group_labels.nunique()
         ).as_hex()
         lut = {}
         legend_elements = []
-        for index, group in enumerate(pd.Series(cond_group_labels).unique()):
+        for index, group in enumerate(cond_group_labels.unique()):
             lut[group] = colors[index]
             legend_elements.append(Patch(color=colors[index], label=group))
-        row_colors = pd.Series(cond_group_labels).map(lut)
+        row_colors = cond_group_labels.map(lut)
         for iii, color in enumerate(row_colors):
             ax.add_patch(
                 plt.Rectangle(
@@ -135,24 +132,19 @@ def plot_gene_factors(data: AnnData, ax: Axes, trim=True):
 
 
 def plot_gene_factors_partial(
-    cmp: int, dataIn: AnnData, ax: Axes, geneAmount: int = 5, top=True
+    cmp: int, dataIn: AnnData, ax: Axes, geneAmount: int = 5
 ):
     """Plotting weights for gene factors for both most negatively/positively weighted terms"""
     cmpName = f"Cmp. {cmp}"
 
     df = pd.DataFrame(
         data=dataIn.varm["Pf2_C"][:, cmp - 1], index=dataIn.var_names, columns=[cmpName]
     )
+    df["abs"] = df[cmpName].abs()
     df = df.reset_index(names="Gene")
-    df = df.sort_values(by=cmpName)
-
-    if top:
-        sns.barplot(
-            data=df.iloc[-geneAmount:, :], x="Gene", y=cmpName, color="k", ax=ax
-        )
-    else:
-        sns.barplot(data=df.iloc[:geneAmount, :], x="Gene", y=cmpName, color="k", ax=ax)
+    df = df.sort_values(by="abs", ascending=False)
 
+    sns.barplot(data=df.iloc[:geneAmount, :], x="Gene", y=cmpName, color="k", ax=ax)
     ax.tick_params(axis="x", rotation=90)
 
 

pf2rnaseq/figures/commonFuncs/plotPaCMAP.py

@@ -62,7 +62,7 @@ def plot_gene_pacmap(gene: str, decompType: str, X: anndata.AnnData, ax: Axes):
 
     # Color by values
     values -= np.min(values)
-    values /= np.max(values)
+    values /= np.max(values) / 10.0
     data["val_cat"] = values
     result = tf.shade(
         agg=canvas.points(data, "x", "y", agg=ds.mean("val_cat")),

pf2rnaseq/figures/figureCITEseq4.py

@@ -13,16 +13,12 @@
 def makeFigure():
     """Get a list of the axis objects and create a figure."""
     # Get list of axis objects
-    ax, f = getSetup((10, 10), (4, 4))
+    ax, f = getSetup((10, 10), (6, 5))
 
-    # Add subplot labels
-    subplotLabel(ax)
-
-    X = read_h5ad("/opt/pf2/CITEseq_fitted_annotated.h5ad", backed="r")
-    comps = [22, 33, 47, 48, 23, 31, 43]
+    X = read_h5ad("/opt/extra-storage/CRC/GSE178341/crc10x_full_50cmp.h5ad", backed="r")
+    comps = range(20, 50)
 
     for i, cmp in enumerate(comps):
-        plot_gene_factors_partial(cmp, X, ax[2 * i], geneAmount=10, top=True)
-        plot_gene_factors_partial(cmp, X, ax[2 * i + 1], geneAmount=10, top=False)
+        plot_gene_factors_partial(cmp, X, ax[i - 1], geneAmount=15)
 
     return f

pf2rnaseq/imports.py

@@ -89,6 +89,17 @@ def import_HTAN() -> anndata.AnnData:
     return prepare_dataset(X, "Condition", geneThreshold=0.1)
 
 
+def import_Vanderbilt_scRNAseq() -> anndata.AnnData:
+    """Imports Vanderbilt's scRNAseq data."""
+
+    nl_epi = anndata.read_h5ad("/opt/extra-storage/CRC/Heiser/VUMC_HTAN_DIS_EPI_V2.h5ad")
+
+    X = anndata.AnnData(X=nl_epi.raw.X, obs=nl_epi.obs, var=nl_epi.raw.var)
+    X.X = csr_matrix(X.X)
+
+    return prepare_dataset(X, "HTAN Specimen ID", geneThreshold=0.01)
+
+
 def import_CCLE() -> anndata.AnnData:
     """Imports barcoded cell data."""
     # TODO: Still need to add gene names and barcodes.