exprmat 0.1.30__tar.gz → 0.1.32__tar.gz

{exprmat-0.1.30 → exprmat-0.1.32}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: exprmat
-Version: 0.1.30
+Version: 0.1.32
 Classifier: Programming Language :: Rust
 Classifier: Programming Language :: Python :: Implementation :: CPython
 Classifier: Programming Language :: Python :: Implementation :: PyPy
@@ -42,7 +42,6 @@ Requires-Dist: metacells
 Requires-Dist: sh
 Requires-Dist: networkx
 Requires-Dist: palantir
-Requires-Dist: scfates
 Requires-Dist: rich
 Requires-Dist: adjusttext
 Requires-Dist: h5py

{exprmat-0.1.30 → exprmat-0.1.32}/exprmat/__init__.py

@@ -8,6 +8,7 @@ import importlib.metadata
 import mudata as mu
 import anndata as ad
 import scanpy as sc
+import pandas as pd
 import pathlib
 
 from exprmat.configuration import default as config

{exprmat-0.1.30 → exprmat-0.1.32}/exprmat/plotting/cnmf.py

@@ -1,6 +1,7 @@
 
 import matplotlib.pyplot as plt
 import pandas as pd
+import numpy as np
 
 
 def cnmf_silhoutte(adata, nmf_slot = 'cnmf', figsize = (3, 2), dpi = 100):
@@ -64,7 +65,11 @@ def cnmf_distance_usages(
     from exprmat.clustering.seriation import serial_matrix, squareform
     serial, order, _ = serial_matrix(squareform(pdistc), method = method)
     indices = adata.obs.iloc[subset, :].iloc[order, :].index.tolist()
-    if annotations: annotations = adata.obs.iloc[subset, :].iloc[order, :][annotations].tolist()
+    if annotations:
+        if annotations in adata.obs.columns: 
+            annotations = adata.obs.iloc[subset, :].iloc[order, :][annotations].tolist()
+        elif annotations in adata.obsm.keys(): 
+            annotations = adata[subset, :][order, :].obsm[annotations]
 
     return matrix_plot(
         serial, labels = indices, annotations = annotations, cmap_annotations = cmap_annotations, 
@@ -90,7 +95,11 @@ def cnmf_distance_modules(
     from exprmat.clustering.seriation import serial_matrix, squareform
     serial, order, _ = serial_matrix(squareform(pdistc), method = method)
     indices = adata.var.iloc[subset, :].iloc[order, :].index.tolist()
-    if annotations: annotations = adata.var.iloc[subset, :].iloc[order, :][annotations].tolist()
+    if annotations: 
+        if annotations in adata.var.columns: 
+            annotations = adata.var.iloc[subset, :].iloc[order, :][annotations].tolist()
+        elif annotations in adata.varm.keys(): 
+            annotations = adata[subset, :][order, :].varm[annotations]
 
     return matrix_plot(
         serial, labels = indices, annotations = annotations, cmap_annotations = cmap_annotations, 
@@ -114,7 +123,8 @@ def matrix_plot(
         ax.set_yticklabels(labels)
         ax.set_xticks([])
 
-    if annotations is not None:
+    # categorical annotations
+    if annotations is not None and isinstance(annotations, list):
         ax_annot = ax.inset_axes([1.02, 0, 0.05, 1], sharey = ax)
         annot_u = list(set(annotations))
         annot_n = [annot_u.index(x) for x in annotations]
@@ -151,4 +161,29 @@ def matrix_plot(
             ncol = legend_cols
         )
     
+    # numerical annotations (from obsm or varm)
+    elif annotations is not None and (
+        isinstance(annotations, np.ndarray) or
+        isinstance(annotations, np.matrix) or
+        isinstance(annotations, pd.DataFrame)
+    ):
+        xn = None
+        if isinstance(annotations, pd.DataFrame):
+            xn = annotations.columns.tolist()
+            annotations = annotations.values
+
+        ncol = annotations.shape[1]
+        ax_annot = ax.inset_axes([1.05, 0, max(0.3, 0.07 * ncol), 1], sharey = ax)
+        
+        ax_annot.imshow(annotations, cmap = cmap_annotations, aspect = 'auto', interpolation = 'nearest')
+        # ax_annot.set_yticks([])
+        if xn: 
+            ax_annot.set_xticks([x for x in range(len(xn))])
+            ax_annot.set_xticklabels(xn, rotation = 90)
+        else:
+            ax_annot.set_xticks([0])
+            ax_annot.set_xticklabels([''])
+
+        ax_annot.tick_params(axis = "y", labelleft = False, length = 0)
+    
     return fig

{exprmat-0.1.30 → exprmat-0.1.32}/exprmat/plotting/de.py

@@ -280,8 +280,8 @@ def marker_plot(
             adata,
             var_names,
             groupby,
-            show=show,
-            save=save,
+            show = False,
+            save = save,
             gene_symbols = gene_symbols,
             **kwds,
         )
@@ -293,8 +293,8 @@ def marker_plot(
             adata,
             var_names,
             groupby,
-            show=show,
-            save=save,
+            show = False,
+            save = save,
             gene_symbols = gene_symbols,
             **kwds,
         )

{exprmat-0.1.30 → exprmat-0.1.32}/exprmat/plotting/expression.py

@@ -8,13 +8,13 @@ from exprmat.utils import find_variable, choose_layer
 
 def barplot(
     adata, *, gene, 
-    slot = 'X', group = 'cell.type', selected_groups = None, 
+    slot = 'X', group = 'cell.type', selected_groups = None, selected_splits = None,
     split = 'treat', palette = ['red', 'black'], ax = None,
     size = (6,3), dpi = 100
 ):
     
-    plt.rcParams["ytick.labelright"] = True
-    plt.rcParams["ytick.labelleft"] = False
+    plt.rcParams["ytick.labelright"] = False
+    plt.rcParams["ytick.labelleft"] = True
     
     colnames = ['logc' + x for x in [gene]]
     cellnames = ['c' + str(i) for i in range(len(adata.obs_names.tolist()))]
@@ -36,6 +36,13 @@ def barplot(
     dropzeros['selected.celltype'] = select_ct
     dropzeros = dropzeros.loc[dropzeros['selected.celltype'],:]
 
+    if selected_splits is None:
+        if split is not None: selected_splits = adata.obs[split].value_counts().index.tolist()
+        else: selected_splits = ['default.split']
+
+    select_sp = [x in selected_splits for x in dropzeros['split'].tolist()]
+    dropzeros = dropzeros.loc[select_sp,:]
+
     if ax is None: fig, axes = plt.subplots(1, 1, figsize = size, dpi = dpi)
     else: axes = ax
 
@@ -49,7 +56,7 @@ def barplot(
     )
     
     axes.tick_params(
-        axis = 'y', right = True, left = False, color = 'gray', length = 4,
+        axis = 'y', right = False, left = True, color = 'gray', length = 4,
         grid_color = 'none'
     )
         
@@ -58,7 +65,7 @@ def barplot(
         grid_color = 'lightgray'
     )
             
-    axes.legend(bbox_to_anchor = (1.2, 1), loc = 'upper left', borderaxespad = 0, frameon = False)
+    axes.legend(bbox_to_anchor = (1, 1), loc = 'upper left', borderaxespad = 0, frameon = False)
     axes.set_xlabel('')
     axes.set_ylabel(gene)
     # axes.set_xticks(selected_groups)
@@ -93,7 +100,7 @@ def barplot(
         axes.set_xticklabels([xtl], rotation = 45)
         pass
 
-    axes.spines[['left', 'top']].set_visible(False)
+    axes.spines[['right', 'top']].set_visible(False)
     axes.yaxis.set_major_formatter(ticker.StrMethodFormatter("{x:.1f}"))
 
     if ax is None: return fig

{exprmat-0.1.30 → exprmat-0.1.32}/exprmat/plotting/lr.py

@@ -81,7 +81,8 @@ def filter_by(liana_res, filter_fun):
     if filter_fun is not None:
         msk = filter_fun(liana_res).astype(bool)
         relevant_interactions = np.unique(liana_res[msk].interaction)
-        liana_res = liana_res[np.isin(liana_res['interaction'], relevant_interactions)]
+        # liana_res = liana_res[np.isin(liana_res['interaction'], relevant_interactions)]
+        liana_res = liana_res[msk]
 
     return liana_res
 
@@ -124,6 +125,7 @@ def pivot_lr(
     
     if mode not in ['counts', 'mean']:
         error("`pivot_mode` must be 'counts' or 'mean'.")
+
     if mode == 'counts':
         pivot_table = liana_res.pivot_table(
             index = source_key, columns = target_key, 
@@ -136,7 +138,7 @@ def pivot_lr(
             values = score_key, aggfunc = 'mean', fill_value = 0
         )
 
-    return pivot_table
+    return pivot_table.loc[pivot_table.columns, :]
 
 
 def scale(arr, min_val = 1, max_val = 5):
@@ -163,15 +165,17 @@ def set_color(adata, label, color_dict = None, hex = True):
             color_dict[x] for x in adata.obs[label].cat.categories
         ]
 
-    elif f"{label}.colors" not in adata.uns:
+    elif f"{label}.colors" not in adata.uns.keys():
         sc.pl._utils._set_default_colors_for_categorical_obs(adata, label)
+        adata.uns[f"{label}.colors"] = adata.uns[f"{label}_colors"]
+        del adata.uns[f"{label}_colors"]
 
     return adata
 
 
 def get_color(adata, label):
 
-    if f"{label}.colors" not in adata.uns: set_color(adata, label)
+    if f"{label}.colors" not in adata.uns.keys(): set_color(adata, label)
     return { 
         x: y for x, y in zip(
             adata.obs[label].cat.categories, 

{exprmat-0.1.30 → exprmat-0.1.32}/exprmat/plotting/utils.py

@@ -2,6 +2,7 @@
 import seaborn as sns
 import numpy as np
 import matplotlib.pyplot as plt
+import pandas as pd
 
 
 def linregress(
@@ -253,4 +254,59 @@ def density(
         axis.set_yticks([yscale(z) for z in yticks])
         if yticklabels: axis.set_yticklabels(yticklabels)
     
+    return fig
+
+
+def waterfall(
+    df: pd.DataFrame, name = 'name', resolution = 'resolution', ax = None,
+    figsize = (3, 3), dpi = 100, ylabel = None
+):
+    if ax is not None: fig, ax = ax.figure, ax
+    else: fig, ax = plt.subplots(1, 1, figsize = figsize, dpi = dpi)
+
+    # sort descending
+    df = df.sort_values([resolution], ascending = False)
+    maximal = 0
+    minimal = 0
+
+    series = df[resolution]
+
+    if (series > 1).sum() > 0: maximal = series.max()
+    elif (series > 0.5).sum() > 0: maximal = 0.8
+    elif (series > 0.3).sum() > 0: maximal = 0.5
+    elif (series > 0).sum() > 0: maximal = 0.3
+    else: maximal = 0
+
+    if (series < 0.8).sum() > 0: minimal = -1
+    elif (series < 0.5).sum() > 0: maximal = -0.8
+    elif (series < 0.3).sum() > 0: maximal = -0.5
+    elif (series < 0).sum() > 0: maximal = -0.3
+    else: minimal = 0
+
+    ticks = [-1, -0.8, -0.5, -0.3, 0, 0.3, 0.5, 0.8, 1, max(1.1, series.max())]
+    ticks = [x for x in ticks if x >= minimal and x <= maximal]
+    hlines = [-0.5, -0.3, 0, 0.3]
+    hlines = [x for x in hlines if x >= minimal and x <= maximal]
+
+    ax.set_xlim(0, len(series))
+    ax.set_ylim(minimal, maximal)
+    ax.set_xticks([])
+    ax.set_yticks(ticks)
+    ax.set_yticklabels([f'{x * 100:.0f}%' for x in ticks])
+    if ylabel: ax.set_ylabel(ylabel)
+
+    for i, val in enumerate(series):
+        color = 'orange'
+        if val > 0.3: color = 'red'
+        elif val < -0.99: color = 'dodgerblue'
+        elif val < -0.3: color = 'gold'
+        ax.fill_between([i + 0.2, i + 0.8], [0, 0], [val, val], color = color)
+    
+    for hl in hlines:
+        ax.hlines(hl, 0, len(series), linestyles = '--', colors = 'red', lw = 1)
+    ax.hlines(0, 0, len(series), linestyles = '-', colors = 'black', lw = 1)
+
+    for pos in ['right', 'top', 'bottom']:
+        ax.spines[pos].set_visible(False)
+
     return fig

{exprmat-0.1.30 → exprmat-0.1.32}/exprmat/reader/experiment.py

@@ -1103,6 +1103,9 @@ class experiment:
     def plot_rna_embedding(self, run_on_samples = False, **kwargs):
         return self.plot_for_rna(run_on_samples, st.rna_plot_embedding, **kwargs)
     
+    def plot_rna_embedding_mask(self, run_on_samples = False, **kwargs):
+        return self.plot_for_rna(run_on_samples, st.rna_plot_embedding_mask, **kwargs)
+    
     def plot_rna_embedding_atlas(self, run_on_samples = False, **kwargs):
         return self.plot_for_rna(run_on_samples, st.rna_plot_embedding_atlas, **kwargs)
 

{exprmat-0.1.30 → exprmat-0.1.32}/exprmat/reader/static.py

@@ -765,6 +765,16 @@ def rna_plot_embedding(adata, sample_name, **kwargs):
     return embedding(adata, sample_name = sample_name, **kwargs)
 
 
+def rna_plot_embedding_mask(adata, sample_name, color, values, **kwargs):
+    from exprmat.reduction.plot import embedding
+    if color in adata.obs.keys():
+        adata.obs['.mask'] = [x in values for x in adata.obs[color]]
+    else: error('`color` must be within adata.obs')
+    fig = embedding(adata, sample_name = sample_name, color = '.mask', **kwargs)
+    if color in adata.obs.keys(): del adata.obs['.mask']
+    return fig
+
+
 def rna_plot_embedding_atlas(adata, sample_name, **kwargs):
     from exprmat.reduction.plot import embedding_atlas
     return embedding_atlas(adata, sample_name = sample_name, **kwargs)
@@ -782,22 +792,37 @@ def rna_plot_gene_gene_regress(adata, sample_name, **kwargs):
 def rna_plot_markers(adata, sample_name, figsize, dpi, **kwargs):
     from exprmat.plotting.de import marker_plot
     pl = marker_plot(adata, sample_name = sample_name, **kwargs)
-    pl.width = figsize[0]
-    pl.height = figsize[1]
-    pl.show()
-    pl.fig.set_dpi(dpi)
-    return pl.fig
+
+    from scanpy.plotting import DotPlot, StackedViolin, MatrixPlot
+    if isinstance(pl, DotPlot) or isinstance(pl, StackedViolin) or isinstance(pl, MatrixPlot):
+        pl.width = figsize[0]
+        pl.height = figsize[1]
+        pl.show()
+        pl.fig.set_dpi(dpi)
+        return pl.fig
+    
+    else:
+        fig = pl['heatmap_ax'].figure if 'heatmap_ax' in pl.keys() else None
+        fig = pl['track_axes'][0].figure if 'track_axes' in pl.keys() else fig
+
+        if fig:
+            fig.set_dpi(dpi)
+            fig.set_figwidth(figsize[0])
+            fig.set_figheight(figsize[1])
+            return fig
+    
 
 
 def rna_plot_expression_bar(
     adata, sample_name, gene, group, split = None,
-    slot = 'X', selected_groups = None, palette = ['red', 'black'], 
+    slot = 'X', selected_groups = None, selected_splits = None, palette = ['red', 'black'], 
     figsize = (6,3), dpi = 100
 ):
     from exprmat.plotting.expression import barplot
     pl = barplot(
         adata, gene = gene, slot = slot, group = group,
-        split = split, selected_groups = selected_groups, palette = palette,
+        split = split, selected_groups = selected_groups, 
+        selected_splits = selected_splits, palette = palette,
         size = figsize, dpi = dpi
     )
     return pl

{exprmat-0.1.30 → exprmat-0.1.32}/exprmat/reduction/plot.py

@@ -655,7 +655,7 @@ def gene_gene(
     from exprmat.utils import find_variable as find_var
     gx = find_var(adata, gene_name = gene_x, layer = layer)
     gy = find_var(adata, gene_name = gene_y, layer = layer)
-    labels = find_var(color) if color is not None else ['.'] * len(gx)
+    labels = find_var(adata, color) if color is not None else ['.'] * len(gx)
 
     df = pd.DataFrame({
         'x': gx,
@@ -749,7 +749,7 @@ def gene_gene(
             )
 
         if legend:
-            assert len(adata.uns[f'{color}.colors']) == len(hue_order)
+            assert len(adata.uns[f'{color}.colors']) >= len(hue_order)
             dummy_objects = []
             legend_artists = {}
             for legend_t, legend_c, legend_id in zip(

{exprmat-0.1.30 → exprmat-0.1.32}/pyproject.toml

@@ -18,7 +18,7 @@ description = "exprmat: Routines for expression matrices"
 readme = "readme.md"
 license = {file = "license"}
 requires-python = ">=3.9, <3.13"
-version = "0.1.30"
+version = "0.1.32"
 authors = [
     {name = "Zheng Yang", email = "xornent@outlook.com"}
 ]
@@ -68,7 +68,6 @@ dependencies = [
     'sh',
     'networkx',
     'palantir',
-    'scfates',
     'rich',
     'adjustText',
     'h5py',