exprmat 0.1.34__tar.gz → 0.1.35__tar.gz

{exprmat-0.1.34 → exprmat-0.1.35}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: exprmat
-Version: 0.1.34
+Version: 0.1.35
 Classifier: Programming Language :: Rust
 Classifier: Programming Language :: Python :: Implementation :: CPython
 Classifier: Programming Language :: Python :: Implementation :: PyPy
@@ -63,6 +63,11 @@ Requires-Dist: scikit-image
 Requires-Dist: spatial-image
 Requires-Dist: multiscale-spatial-image
 Requires-Dist: tangram-sc
+Requires-Dist: shapely>=2.0.0
+Requires-Dist: pyogrio>=0.7.2
+Requires-Dist: pyproj>=3.3.0
+Requires-Dist: geopandas
+Requires-Dist: packaging
 Requires-Dist: sphinx ; extra == 'dev'
 Requires-Dist: twine ; extra == 'dev'
 Requires-Dist: ipykernel ; extra == 'dev'

{exprmat-0.1.34 → exprmat-0.1.35}/exprmat/__init__.py

@@ -10,12 +10,13 @@ import anndata as ad
 import scanpy as sc
 import pandas as pd
 import pathlib
+from PIL import Image
 
 from exprmat.configuration import default as config
 from exprmat.ansi import warning, info
 
 mu.set_options(pull_on_update = False)
-
+Image.MAX_IMAGE_PIXELS = config['max.image']
 
 DATABASE_SETUP_MESSAGE = """
 The exprmat package do not come with a included database after version 0.1.25, 

{exprmat-0.1.34 → exprmat-0.1.35}/exprmat/configuration.py

@@ -38,6 +38,8 @@ class Configuration:
             'hsa': 'grch38'
         }
 
+        self.config['max.image'] = int(10000 * 50000)
+
 
     def __getitem__(self, index):
         return self.config[index]

{exprmat-0.1.34 → exprmat-0.1.35}/exprmat/plotting/expression.py

@@ -3,14 +3,16 @@ import seaborn as sns
 from matplotlib import ticker
 import matplotlib.pyplot as plt
 import pandas as pd
+
 from exprmat.utils import find_variable, choose_layer
+from exprmat.ansi import error
 
 
 def barplot(
     adata, *, gene, 
     slot = 'X', group = 'cell.type', selected_groups = None, selected_splits = None,
-    split = 'treat', palette = ['red', 'black'], ax = None,
-    size = (6,3), dpi = 100
+    split = 'treat', palette = ['red', 'black'], ax = None, style = 'box',
+    size = (6,3), dpi = 100, violin_kwargs = {}
 ):
     
     plt.rcParams["ytick.labelright"] = False
@@ -46,14 +48,28 @@ def barplot(
     if ax is None: fig, axes = plt.subplots(1, 1, figsize = size, dpi = dpi)
     else: axes = ax
 
-    sns.boxplot(
-        data = dropzeros.loc[dropzeros['name'] == gene,:],
-        x = "row", y = "logc", hue = 'split', ax = axes, fill = True,
-        palette = palette, saturation = 0.8, log_scale = False,
-        color = '.8', linecolor = '0', order = selected_groups, gap = 0.3,
-        flierprops = dict(markerfacecolor = '0.1', markersize = 1, linestyle = 'none'),
-        showfliers = True, width = 0.6
-    )
+    if style == 'box':
+
+        sns.boxplot(
+            data = dropzeros.loc[dropzeros['name'] == gene,:],
+            x = "row", y = "logc", hue = 'split', ax = axes, fill = True,
+            palette = palette, saturation = 0.8, log_scale = False,
+            color = '.8', linecolor = '0', order = selected_groups, gap = 0.3,
+            flierprops = dict(markerfacecolor = '0.1', markersize = 1, linestyle = 'none'),
+            showfliers = True, width = 0.6
+        )
+    
+    elif style == 'violin':
+
+        sns.violinplot(
+            data = dropzeros.loc[dropzeros['name'] == gene,:],
+            x = "row", y = "logc", hue = 'split', ax = axes, fill = True,
+            palette = palette, saturation = 0.8, log_scale = False,
+            color = '.8', linecolor = '0', order = selected_groups, gap = 0,
+            width = 0.8, **violin_kwargs
+        )
+    
+    else: error('can only set plot style to `box` or `violin`.')
     
     axes.tick_params(
         axis = 'y', right = False, left = True, color = 'gray', length = 4,

{exprmat-0.1.34 → exprmat-0.1.35}/exprmat/reader/experiment.py

@@ -166,14 +166,54 @@ class experiment:
                 
                 if not 'rnasp-c' in self.modalities.keys(): self.modalities['rnasp-c'] = {}
 
-                from exprmat.reader.spatial import read_seekspace
-                self.modalities['rnasp-c'][i_sample] = read_seekspace(
+                from exprmat.reader.spatial import read_seekspace, read_xenium_explorer, is_xenium_explorer
+                
+                if is_xenium_explorer(i_loc):
+                    self.modalities['rnasp-c'][i_sample] = read_xenium_explorer(
+                        src = i_loc, prefix = '', metadata = meta, sample = i_sample,
+                        raw = False, default_taxa = i_taxa, eccentric = eccentric
+                    )
+                
+                else:
+                    self.modalities['rnasp-c'][i_sample] = read_seekspace(
+                        src = i_loc, prefix = '', metadata = meta, sample = i_sample,
+                        raw = False, default_taxa = i_taxa, eccentric = eccentric
+                    )
+
+                self.modalities['rnasp-c'][i_sample].var = \
+                    st.search_genes(self.modalities['rnasp-c'][i_sample].var_names.tolist())
+                
+            elif i_mod == 'rnasp-b':
+                
+                if not 'rnasp-b' in self.modalities.keys(): self.modalities['rnasp-b'] = {}
+
+                from exprmat.reader.spatial import read_visium
+                self.modalities['rnasp-b'][i_sample] = read_visium(
                     src = i_loc, prefix = '', metadata = meta, sample = i_sample,
                     raw = False, default_taxa = i_taxa, eccentric = eccentric
                 )
 
-                self.modalities['rnasp-c'][i_sample].var = \
-                    st.search_genes(self.modalities['rnasp-c'][i_sample].var_names.tolist())
+                self.modalities['rnasp-b'][i_sample].var = \
+                    st.search_genes(self.modalities['rnasp-b'][i_sample].var_names.tolist())
+                
+            elif i_mod == 'rnasp-s':
+                
+                if not 'rnasp-s' in self.modalities.keys(): self.modalities['rnasp-s'] = {}
+
+                from exprmat.reader.spatial import read_visium_hd
+                self.modalities['rnasp-s'][i_sample], cellseg = read_visium_hd(
+                    src = i_loc, prefix = '', metadata = meta, sample = i_sample,
+                    raw = False, default_taxa = i_taxa, eccentric = eccentric
+                )
+
+                self.modalities['rnasp-s'][i_sample].var = \
+                    st.search_genes(self.modalities['rnasp-s'][i_sample].var_names.tolist())
+                
+                if cellseg is not None:
+                    if not 'rnasp-c' in self.modalities.keys(): self.modalities['rnasp-c'] = {}
+                    self.modalities['rnasp-c'][i_sample] = cellseg
+                    self.modalities['rnasp-c'][i_sample].var = \
+                        st.search_genes(self.modalities['rnasp-c'][i_sample].var_names.tolist())
             
             elif i_mod == 'atac':
                 
@@ -570,7 +610,7 @@ class experiment:
         else: self.mudata = None
 
 
-    def do_for(self, samples, func, **kwargs):
+    def do_for(self, modality, samples, func, **kwargs):
         
         results = {}
         for mod, samp in zip(
@@ -578,6 +618,7 @@ class experiment:
             self.metadata.dataframe['sample'].tolist()
         ):
             if '.' in mod: continue
+            if mod != modality: continue
 
             do = False
             if samples is None: do = True
@@ -608,9 +649,13 @@ class experiment:
 
     def do_for_modality(self, modality, run_on_samples, func, **kwargs):
         if isinstance(run_on_samples, bool) and run_on_samples:
-            return self.do_for(self.all_samples(modality), func, **kwargs)
+            return self.do_for(modality, self.all_samples(modality), func, **kwargs)
         elif isinstance(run_on_samples, list):
-            return self.do_for(list(set(self.all_samples(modality)) & set(run_on_samples)), func, **kwargs)
+            return self.do_for(
+                modality, list(set(self.all_samples(modality)) & set(run_on_samples)), 
+                func, **kwargs
+            )
+        
         else:
             assert modality in self.mudata.mod.keys()
             return func(self.mudata[modality], 'integrated', **kwargs)
@@ -629,6 +674,9 @@ class experiment:
     
     def do_for_rnaspc(self, run_on_samples, func, **kwargs):
         return self.do_for_modality('rnasp-c', run_on_samples, func, **kwargs)
+    
+    def do_for_rnaspb(self, run_on_samples, func, **kwargs):
+        return self.do_for_modality('rnasp-b', run_on_samples, func, **kwargs)
         
     
     def plot_for_modality(
@@ -639,12 +687,16 @@ class experiment:
         setup_styles()
 
         if isinstance(run_on_samples, bool) and run_on_samples:
-            figures = self.do_for(self.all_samples(modality), func, **kwargs)
+            figures = self.do_for(modality, self.all_samples(modality), func, **kwargs)
             for f in figures.values(): f.tight_layout()
             return figures
         
         elif isinstance(run_on_samples, list):
-            figures = self.do_for(list(set(self.all_samples(modality)) & set(run_on_samples)), func, **kwargs)
+            figures = self.do_for(
+                modality, list(set(self.all_samples(modality)) & set(run_on_samples)), 
+                func, **kwargs
+            )
+            
             for f in figures.values(): f.tight_layout()
             return figures
         
@@ -727,6 +779,15 @@ class experiment:
             'rnasp-c', run_on_samples, func, 
             run_on_splits, split_key, split_selection, **kwargs
         )
+    
+    def plot_for_rnaspb(
+        self, run_on_samples, func,
+        run_on_splits = False, split_key = None, split_selection = None, **kwargs
+    ):
+        return self.plot_for_modality(
+            'rnasp-b', run_on_samples, func, 
+            run_on_splits, split_key, split_selection, **kwargs
+        )
 
     
     def all_samples(self, modality = None):
@@ -1337,6 +1398,32 @@ class experiment:
         '''
         return self.do_for_rnaspc(run_on_samples, st.rna_proportion, **kwargs)
 
+    def run_rnaspc_roi(self, run_on_samples = False, **kwargs):
+        
+        roi = self.do_for_rnaspc(run_on_samples, st.rnaspc_roi, **kwargs)
+        if isinstance(roi, dict):
+            roi = roi[kwargs.get('spsample')]
+        
+        roi.uns['spatial'] = {
+            kwargs.get('sample_added'): roi.uns['spatial'][kwargs.get('spsample')]
+        }
+
+        rows = self.metadata.dataframe[(
+            (self.metadata.dataframe['sample'] == kwargs.get('spsample')) &
+            (self.metadata.dataframe['modality'] == 'rnasp-c')
+        )]
+
+        assert len(rows) == 1
+        props = rows.iloc[0].copy()
+        props['sample'] = kwargs.get('sample_added')
+        props['modality'] = 'rnasp-c'
+
+        self.metadata.insert_row(props)
+        if not 'rnasp-c' in self.modalities.keys(): self.modalities['rnasp-c'] = {}
+        self.modalities['rnasp-c'][kwargs.get('sample_added')] = roi
+        self.modalities['rnasp-c'][kwargs.get('sample_added')].var = \
+            st.search_genes(self.modalities['rnasp-c'][kwargs.get('sample_added')].var_names.tolist())
+
 
     # plotting wrappers
 
@@ -1503,6 +1590,10 @@ class experiment:
         return self.plot_for_rnaspc(run_on_samples, st.rnaspc_plot_embedding_spatial, **kwargs)
     
 
+    def plot_rnaspb_embedding_spatial(self, run_on_samples = False, **kwargs):
+        return self.plot_for_rnaspb(run_on_samples, st.rnaspc_plot_embedding_spatial, **kwargs)
+    
+
     def plot_sankey(self, run_on_samples = False, **kwargs):
         return self.plot_for_rna(run_on_samples, st.adata_plot_sankey, **kwargs)
     
@@ -1678,6 +1769,17 @@ class experiment:
         return merge
     
 
+    # magic accessors
+
+    def __getitem__(self, key):
+        self.check_merged()
+        if key in self.mudata.mod.keys():
+            return self.mudata[key]
+        else: 
+            warning(f'key must be one of [{", ".join(list(self.mudata.mod.keys()))}]')
+            error(f'no integrated modality named `{key}`.')
+            
+
     def __repr__(self):
 
         from exprmat.ansi import green, cyan, red, yellow

{exprmat-0.1.34 → exprmat-0.1.35}/exprmat/reader/matcher.py

@@ -250,7 +250,7 @@ def adjust_features(path, refine_finder = False, default_taxa = 'mmu', eccentric
 def read_mtx_rna(
     src: str, prefix: str, 
     metadata: metadata, sample: str, raw: bool = False,
-    default_taxa = 'mmu', eccentric = None
+    default_taxa = 'mmu', eccentric = None, suppress_filter = False,
 ):
     '''
     Formalize the three components in the matrix directory, and read the folder with gene
@@ -299,7 +299,41 @@ def read_mtx_rna(
     #     return None
 
     final = match_matrix_rna(
-        adata, metadata, sample, 
+        adata, metadata, sample, suppress_filter = suppress_filter,
+        force_filter = raw, default_taxa = default_taxa
+    )
+
+    warnings.filterwarnings('default')
+    
+    del adata
+    return final
+
+
+def read_h5_rna(
+    path: str, 
+    metadata: metadata, sample: str, raw: bool = False,
+    default_taxa = 'mmu', eccentric = None, suppress_filter = False,
+):
+    import warnings
+    warnings.filterwarnings('ignore')
+
+    adata = sc.read_10x_h5(
+        path, gex_only = True
+    )
+
+    adata.var_names = adata.var['gene_ids'].tolist()
+    del adata.var
+
+    if eccentric is not None:
+        adata.var_names = [eccentric(x) for x in adata.var_names]
+
+    # except Exception as ex: 
+    #     warning('error occurs when reading matrix files:')
+    #     warning(str(ex))
+    #     return None
+
+    final = match_matrix_rna(
+        adata, metadata, sample, suppress_filter = suppress_filter,
         force_filter = raw, default_taxa = default_taxa
     )
 
@@ -429,24 +463,27 @@ def read_h5ad_rna(
 
 
 def match_matrix_rna(
-        adata, metadata: metadata, sample: str, 
-        force_filter = False, default_taxa = 'mmu',
-        do_not_alter_obs_names = False
-    ):
+    adata, metadata: metadata, sample: str, 
+    force_filter = False, default_taxa = 'mmu',
+    do_not_alter_obs_names = False, suppress_filter = False
+):
 
     # if more than 50000 cells in a single matrix, we just believe that it is
     # an unfiltered raw matrix. we should roughly filter the empty droplets
 
-    if adata.n_obs > 50000 or force_filter:
+    if (adata.n_obs > 50000 or force_filter) and (not suppress_filter):
         valid = (adata.X.sum(axis = 1).transpose() > 200).tolist()[0]
         adata_f = adata[valid, :]
+
     else: adata_f = adata
 
     rows = metadata.dataframe[metadata.dataframe['sample'] == sample]
     assert len(rows) >= 1
     rows = rows[
         (rows['modality'] == 'rna') |
-        (rows['modality'] == 'rnasp-c')
+        (rows['modality'] == 'rnasp-c') |
+        (rows['modality'] == 'rnasp-b') |
+        (rows['modality'] == 'rnasp-s')
     ]
     assert len(rows) == 1
     props = rows.iloc[0]

{exprmat-0.1.34 → exprmat-0.1.35}/exprmat/reader/metadata.py

@@ -237,6 +237,40 @@ class metadata:
         temp = [value if (contains in str(x)) else y for x, y in zip(source, temp)]
         self.dataframe[dest] = temp
 
+
+    def find(self, modality, sample):
+        
+        from exprmat.ansi import error, warning
+        df = self.dataframe[(
+            (self.dataframe['sample'] == sample) &
+            (self.dataframe['modality'] == modality)
+        )]
+
+        if len(df) == 1:
+            return df.iloc[0]
+        elif len(df) == 0:
+            warning(f'could not find [{modality}] {sample}')
+            return None
+        else: error(f'duplicated sample for {sample}')
+
+
+    def insert_row(self, row):
+
+        from exprmat.ansi import error
+        insert_index = len(self.dataframe)
+        # get the modality and sample name if there exists one:
+        selection: pandas.Series = (
+            (self.dataframe['modality'] == row['modality']) &
+            (self.dataframe['sample'] == row['sample'])
+        )
+
+        if selection.sum() == 1:
+            insert_index = selection.tolist().index(True)
+        elif selection.sum() == 0: pass
+        else: error('failed to insert row for duplicated index.')
+
+        self.dataframe.loc[insert_index] = row
+
     
 def load_metadata(fpath):
     '''