exprmat 0.1.10__tar.gz → 0.1.12__tar.gz

{exprmat-0.1.10 → exprmat-0.1.12}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: exprmat
-Version: 0.1.10
+Version: 0.1.12
 Summary: routines to process expression matrices
 Home-page: https://github.com/yang-z-03/exprmat
 Author: Zheng Yang

exprmat-0.1.12/exprmat/__init__.py

@@ -0,0 +1,7 @@
+
+import mudata as mu
+import anndata as ad
+
+mu.set_options(pull_on_update = False)
+
+def version(): print('exprmat 0.1.12 [revision 1] [dev]')

exprmat-0.1.12/exprmat/descriptive/aggregate.py

@@ -0,0 +1,127 @@
+'''
+Summary accepts two to three categoricals in the obs metadata to reform the dataset
+with summarizing statistics. Aggregates takes one obs category and one var category
+to aggregate observations and variables.
+'''
+
+import numpy as np
+import anndata as ad
+
+from exprmat.utils import choose_layer
+
+
+def aggregate(
+    adata, data = 'X', method = 'mean', method_args = {},
+    obs_key = 'sample', var_key = None
+):
+    source = choose_layer(adata, layer = data)
+
+    # TODO: maybe some alternative method rather than silly iteration :(
+
+    if (obs_key is None) and (var_key is not None):
+        return aggregate_var(
+            adata, data = data, method = method, method_args = method_args,
+            var_key = var_key
+        )
+    
+    if (var_key is None) and (obs_key is not None):
+        return aggregate_obs(
+            adata, data = data, method = method, method_args = method_args,
+            obs_key = obs_key
+        )
+    
+    obs_val = adata.obs_names.tolist()
+    obs = adata.obs_names.to_numpy()
+    if obs_key is not None:
+        obs_val = adata.obs[obs_key].unique().tolist()
+        obs = adata.obs[obs_key].to_numpy()
+    
+    var_val = adata.var_names.tolist()
+    var = adata.var_names.to_numpy()
+    if var_key is not None:
+        var_val = adata.var[var_key].unique().tolist()
+        var = adata.var[var_key].to_numpy()
+    
+    mat = np.ndarray(shape = (len(obs_val), len(var_val)), dtype = np.float32)
+    meth = None
+
+    methods = {
+        'mean': np.mean
+    }
+
+    for i in range(len(obs_val)):
+        for j in range(len(var_val)):
+            
+            mat[i, j] = methods[method](
+                source[obs == obs_val[i], var == var_val[j]], 
+                **method_args
+            )
+    
+    annd = ad.AnnData(X = mat)
+    annd.obs_names = obs_val
+    annd.var_names = var_val
+    return annd
+
+
+def aggregate_obs(
+    adata, data = 'X', method = 'mean', method_args = {}, obs_key = 'sample'
+):
+    
+    source = choose_layer(adata, layer = data)
+
+    obs_val = adata.obs_names.tolist()
+    obs = adata.obs_names.to_numpy()
+    if obs_key is not None:
+        obs_val = adata.obs[obs_key].unique().tolist()
+        obs = adata.obs[obs_key].to_numpy()
+    
+    mat = np.ndarray(shape = (len(obs_val), adata.n_vars), dtype = np.float32)
+
+    methods = {
+        'mean': np.mean
+    }
+
+    for i in range(len(obs_val)):
+        mat[i, :] = methods[method](
+            source[obs == obs_val[i], :],
+            axis = 0,
+            **method_args
+        )
+    
+    annd = ad.AnnData(X = mat)
+    annd.obs_names = obs_val
+    annd.var_names = adata.var_names.tolist()
+    annd.var = adata.var.copy()
+    return annd
+
+
+def aggregate_var(
+    adata, data = 'X', method = 'mean', method_args = {}, var_key = 'module'
+):
+    
+    source = choose_layer(adata, layer = data)
+
+    var_val = adata.var_names.tolist()
+    var = adata.var_names.to_numpy()
+    if var_key is not None:
+        var_val = adata.var[var_key].unique().tolist()
+        var = adata.var[var_key].to_numpy()
+    
+    mat = np.ndarray(shape = (adata.n_obs, len(var_val)), dtype = np.float32)
+
+    methods = {
+        'mean': np.mean
+    }
+
+    for i in range(len(var_val)):
+        mat[:, i] = methods[method](
+            source[:, var == var_val[i]],
+            axis = 1, # row sums
+            **method_args
+        )
+    
+    annd = ad.AnnData(X = mat)
+    annd.obs_names = adata.obs_names.tolist()
+    annd.obs = adata.obs.copy()
+    annd.var_names = var_val
+    return annd

{exprmat-0.1.10/exprmat/clustering → exprmat-0.1.12/exprmat/descriptive}/summary.py

@@ -22,8 +22,12 @@ from exprmat.utils import choose_layer
 
 def summarize(
     adata, data = 'X', method = 'n', method_args = {},
-    orient = 'obs', on = 'sample', across = None,
+    
+    orient = 'obs', 
+    on = 'sample', 
+    across = None,
     split = None, 
+
     attached_metadata_on = None, attached_metadata_across = None,
     attach_method_on = 'first', attach_method_across = 'first'
 ):

{exprmat-0.1.10 → exprmat-0.1.12}/exprmat/plotting/lr.py

@@ -79,7 +79,7 @@ def invert_scores(score):
 
 def filter_by(liana_res, filter_fun):
     if filter_fun is not None:
-        msk = liana_res.apply(filter_fun, axis=1).astype(bool)
+        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)]
 
@@ -428,14 +428,20 @@ def lr_dotplot(
     # ax.set_facecolor('lightgrey')
     # plt.grid(True, zorder = -1)
 
-    ax.scatter(
+    p = ax.scatter(
         x = 'x', y = 'y', s = 's',
-        edgecolors = "none", c = 'c',
+        edgecolors = "lightgrey", c = 'c', cmap = cmap,
         data = pd.DataFrame(df), marker = 'o'
     )
 
-    for label in ax.get_xticklabels(): label.set_rotation(90)
+    cbar = plt.colorbar(
+        p, shrink = 0.3, aspect = 10,
+        anchor = (0.2, 0), location = "right"
+    )
 
+    cbar.ax.set_title(colour, loc = "left", fontsize = 10, pad = 15)
+
+    for label in ax.get_xticklabels(): label.set_rotation(90)
     return fig
 
 

{exprmat-0.1.10 → exprmat-0.1.12}/exprmat/plotting/summary.py

@@ -4,7 +4,7 @@ import pandas as pd
 import numpy as np
 import matplotlib.pyplot as plt
 
-from exprmat.clustering.summary import to_long_form
+from exprmat.descriptive.summary import to_long_form
 from exprmat.ansi import error, info
 
 

{exprmat-0.1.10 → exprmat-0.1.12}/exprmat/reader/experiment.py

@@ -822,7 +822,7 @@ class experiment:
         attached_metadata_on = None, attached_metadata_across = None,
         attach_method_on = 'first', attach_method_across = 'first'
     ):
-        from exprmat.clustering.summary import summarize
+        from exprmat.descriptive.summary import summarize
         return summarize(
             adata, data = data, method = method, method_args = method_args,
             orient = orient, on = on, across = across, split = split,
@@ -831,6 +831,18 @@ class experiment:
             attach_method_on = attach_method_on,
             attach_method_across = attach_method_across
         )
+    
+
+    @staticmethod
+    def rna_aggregate(
+        adata, sample_name, data = 'X', method = 'mean', method_args = {},
+        obs_key = 'sample', var_key = None
+    ):
+        from exprmat.descriptive.aggregate import aggregate
+        return aggregate(
+            adata, data = data, method = method, method_args = method_args,
+            obs_key = obs_key, var_key = var_key
+        )
 
 
     @staticmethod
@@ -1399,27 +1411,27 @@ class experiment:
 
     @staticmethod
     def rna_plot_gsea_dotplot(
-        adata, sample_name, gsea_key, max_fdr = 1, max_p = 0.05,
+        adata, sample_name, gsea_key, max_fdr = 1, max_p = 0.05, top_term: int = 100,
         colour = 'p', title = "", cmap = 'turbo', figsize = (3, 2), cutoff = 1, ptsize = 5
     ):
         from exprmat.plotting.gse import gsea_dotplot
         return gsea_dotplot(
             experiment.rna_get_gsea(adata, gsea_key, max_fdr = max_fdr, max_p = max_p),
             column = colour, x = 'nes', y = 'name', title = gsea_key if title is None else title,
-            cmap = cmap, size = ptsize, figsize = figsize, cutoff = cutoff
+            cmap = cmap, size = ptsize, figsize = figsize, cutoff = cutoff, top_term = top_term
         )
     
 
     @staticmethod
     def rna_plot_opa_dotplot(
-        adata, sample_name, opa_key, max_fdr = 1, max_p = 0.05,
+        adata, sample_name, opa_key, max_fdr = 1, max_p = 0.05, top_term: int = 100,
         colour = 'fdr', title = None, cmap = 'turbo', figsize = (3, 2), cutoff = 1, ptsize = 5
     ):
         from exprmat.plotting.gse import opa_dotplot
         return opa_dotplot(
             experiment.rna_get_opa(adata, opa_key, max_fdr = max_fdr, max_p = max_p),
             column = colour, x = 'or', y = 'term', title = opa_key if title is None else title,
-            cmap = cmap, size = ptsize, figsize = figsize, cutoff = cutoff
+            cmap = cmap, size = ptsize, figsize = figsize, cutoff = cutoff, top_term = top_term
         )
     
 
@@ -1615,6 +1627,9 @@ class experiment:
     def run_rna_summary(self, run_on_samples = False, **kwargs):
         return self.do_for_rna(run_on_samples, experiment.rna_summary, **kwargs)
     
+    def run_rna_aggregate(self, run_on_samples = False, **kwargs):
+        return self.do_for_rna(run_on_samples, experiment.rna_aggregate, **kwargs)
+    
     def run_rna_attach_tcr(self, run_on_samples = False):
         return self.do_for_rna(
             run_on_samples, 

{exprmat-0.1.10 → exprmat-0.1.12}/exprmat.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: exprmat
-Version: 0.1.10
+Version: 0.1.12
 Summary: routines to process expression matrices
 Home-page: https://github.com/yang-z-03/exprmat
 Author: Zheng Yang

{exprmat-0.1.10 → exprmat-0.1.12}/exprmat.egg-info/SOURCES.txt

@@ -13,7 +13,6 @@ exprmat/clustering/__init__.py
 exprmat/clustering/leiden.py
 exprmat/clustering/rogue.py
 exprmat/clustering/spectral.py
-exprmat/clustering/summary.py
 exprmat/clustering/utils.py
 exprmat/cnv/__init__.py
 exprmat/cnv/cnvscore.py
@@ -109,9 +108,11 @@ exprmat/data/mmu/genesets/transcript-factors.gmt
 exprmat/data/mmu/genesets/wikipathway.gmt
 exprmat/data/mmu/lr/consensus
 exprmat/descriptive/__init__.py
+exprmat/descriptive/aggregate.py
 exprmat/descriptive/de.py
 exprmat/descriptive/gse.py
 exprmat/descriptive/kde.py
+exprmat/descriptive/summary.py
 exprmat/lr/__init__.py
 exprmat/lr/aggregate.py
 exprmat/lr/method.py

{exprmat-0.1.10 → exprmat-0.1.12}/setup.py

@@ -3,7 +3,7 @@ from setuptools import setup, find_packages
 
 setup(
     name                 = 'exprmat',
-    version              = '0.1.10',
+    version              = '0.1.12',
     description          = 'routines to process expression matrices',
     url                  = 'https://github.com/yang-z-03/exprmat',
     author               = 'Zheng Yang',

exprmat-0.1.10/exprmat/__init__.py

@@ -1,5 +0,0 @@
-
-import mudata as mu
-import anndata as ad
-
-mu.set_options(pull_on_update = False)