maradoner 0.14.1__tar.gz → 0.14.3__tar.gz

{maradoner-0.14.1 → maradoner-0.14.3}/PKG-INFO

@@ -1,17 +1,16 @@
 Metadata-Version: 2.1
 Name: maradoner
-Version: 0.14.1
+Version: 0.14.3
 Summary: Variance-adjusted estimation of motif activities.
-Home-page: https://github.com/autosome-ru/nemara
+Home-page: https://github.com/autosome-ru/maradoner
 Author: Georgy Meshcheryakov
 Author-email: iam@georgy.top
-Classifier: Programming Language :: Python :: 3.9
-Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Scientific/Engineering
 Classifier: Operating System :: OS Independent
-Requires-Python: >=3.9
+Requires-Python: >=3.11
 Description-Content-Type: text/markdown
 Requires-Dist: pip>=24.0
 Requires-Dist: typer>=0.13
@@ -27,6 +26,7 @@ Requires-Dist: dill>=0.3.9
 Requires-Dist: rich>=12.6.0
 Requires-Dist: tqdm>=4.0
 Requires-Dist: scikit-learn>=1.6
+Requires-Dist: tables>=3.10
 
 
 **MARADONER**

{maradoner-0.14.1 → maradoner-0.14.3}/maradoner/__init__.py

@@ -1,5 +1,5 @@
 # -*- coding: utf-8 -*-
-__version__ = '0.14.1'
+__version__ = '0.14.3'
 import importlib
 
 
@@ -17,7 +17,8 @@ __min_reqs__ = [
             'dill>=0.3.9',
             'rich>=12.6.0',
             'tqdm>=4.0',
-            'scikit-learn>=1.6'
+            'scikit-learn>=1.6',
+            'tables>=3.10'
            ]
 
 def versiontuple(v):

{maradoner-0.14.1 → maradoner-0.14.3}/maradoner/create.py

@@ -13,7 +13,7 @@ def transform_loadings(df, mode: str, zero_cutoff=1e-9, prom_inds=None):
         df[df < zero_cutoff] = 0
         df = (df - df.min(axis=None)) / (df.max(axis=None) - df.min(axis=None))
     stds = df.std()
-    drop_inds = (stds < 1e-16) | np.isnan(stds)
+    drop_inds = (stds == 0) | np.isnan(stds)
     if prom_inds is not None:
         df = df.loc[prom_inds, ~drop_inds]
     else:

{maradoner-0.14.1 → maradoner-0.14.3}/maradoner/grn.py

@@ -3,12 +3,11 @@ import numpy as np
 import jax.numpy as jnp
 import jax 
 from .utils import read_init, openers, ProjectData
-from .fit import FOVResult, ActivitiesPrediction, FitResult
-from scipy.optimize import minimize_scalar, minimize
+from .fit import ActivitiesPrediction, FitResult
+from scipy.optimize import minimize
 import os
 import dill
 from pandas import DataFrame as DF
-from scipy.stats import norm
 from functools import partial
 from tqdm import tqdm
 
@@ -20,6 +19,8 @@ def estimate_promoter_prior_variance(data: ProjectData, activities: ActivitiesPr
     group_inds = data.group_inds
     Y = Y - fit.promoter_mean.mean.reshape(-1, 1) - fit.sample_mean.mean.reshape(1, -1)
     Y = Y -  B @ fit.motif_mean.mean.reshape(-1, 1)
+    if activities.filtered_motifs is not None and len(activities.filtered_motifs):
+        B = np.delete(B, activities.filtered_motifs, axis=1)
     Y = np.concatenate([Y[:, inds].mean(axis=1, keepdims=True) - B @ U.reshape(-1, 1)
                         for inds, U in zip(group_inds, activities.U.T)],
                        axis=1)
@@ -53,13 +54,14 @@ def estimate_promoter_variance(project_name: str, prior_top=0.90):
                                                  top=prior_top)
     print('Piror standard deviation:', prior_var ** 0.5)
     prior_means = fit.error_variance.variance
-    
     Y = Y - fit.promoter_mean.mean.reshape(-1, 1) - fit.sample_mean.mean.reshape(1, -1)
     Y = Y - B @ fit.motif_mean.mean.reshape(-1, 1)
     Y = Y ** 2
     B_hat = B ** 2 * fit.motif_variance.motif
     B_hat = B_hat.sum(axis=1)
     var = list()
+    jax.config.update('jax_enable_x64', True)
+    jax.config.update('jax_platform_name', 'cpu') 
     for inds, prior_mean, nu in tqdm(list(zip(group_inds, prior_means, fit.motif_variance.group))):
         Yt = Y[:, inds].sum(axis=1)
         s = len(inds)
@@ -78,6 +80,47 @@ def estimate_promoter_variance(project_name: str, prior_top=0.90):
     return var
     
 
+def bayesian_fdr_control(p0, alpha=0.05):
+    """
+    Control Bayesian FDR using sorted posterior probabilities of H0.
+    
+    Args:
+        p0: Array of posterior probabilities P(H0|X) for each hypothesis.
+        alpha: Target FDR level (e.g., 0.05).
+    
+    Returns:
+        discoveries: Boolean array (True = reject H0).
+        threshold: Rejection threshold for P(H0|X).
+    """
+    p0 = np.asarray(p0)
+    if p0.size == 0:
+        return np.zeros_like(p0, dtype=bool), -np.inf
+    
+    # Sort in ascending order
+    sorted_p0 = np.sort(p0)
+    m = len(sorted_p0)
+    
+    # Compute cumulative FDR = (sum_{i=1}^k p_i) / k
+    cum_sum = np.cumsum(sorted_p0)
+    fdr_k = cum_sum / np.arange(1, m + 1)
+    
+    # Find largest k where FDR(k) <= alpha
+    valid_indices = np.where(fdr_k <= alpha)[0]  # 0-based indices
+    if len(valid_indices) == 0:
+        k = 0
+    else:
+        k = valid_indices[-1] + 1  # Convert to 1-based index
+    
+    # Set threshold and discover
+    if k > 0:
+        threshold = sorted_p0[k - 1]  # k-th element (0-indexed at k-1)
+        discoveries = p0 <= threshold
+    else:
+        threshold = -np.inf
+        discoveries = np.zeros_like(p0, dtype=bool)
+    
+    return discoveries, threshold
+
 def grn(project_name: str,  output: str, use_hdf=False, save_stat=True,
         fdr_alpha=0.05, prior_h1=1/100):
     data = read_init(project_name)
@@ -153,11 +196,20 @@ def grn(project_name: str,  output: str, use_hdf=False, save_stat=True,
         lr = lr[inds]
         belief = belief[inds]
         belief = belief.astype(np.half)
-        sorted_beliefs = np.sort(belief.flatten())
-        cumulative_fdr = np.cumsum(sorted_beliefs) / (np.arange(len(sorted_beliefs)) + 1)
+        # fdr_threshold = bayesian_fdr_control(belief.flatten(), fdr_alpha)
+        # print(fdr_threshold[0].mean(), fdr_threshold[1])
+        # fdr_threshold = fdr_threshold[1]
+        sorted_beliefs = belief.flatten() 
+        sorted_beliefs = sorted_beliefs[sorted_beliefs > 0.5]
+        sorted_beliefs = np.sort(sorted_beliefs)[::-1]
+        sorted_inbeliefs = 1 - sorted_beliefs
+        
+        cumulative_fdr = np.cumsum(sorted_inbeliefs) / (np.arange(len(sorted_inbeliefs)) + 1)
+        # print(fdr_alpha)
         try:
-            k = np.max(np.where(cumulative_fdr <= fdr_alpha)[0])
-            fdr_threshold = sorted_beliefs[k-1]
+            k = np.min(np.where(cumulative_fdr <= fdr_alpha)[0])
+            fdr_threshold = sorted_beliefs[k]
+            print(k, fdr_threshold)
         except ValueError:
             fdr_threshold = 1.0
         filename = os.path.join(folder_belief, f'{name}.txt')

{maradoner-0.14.1 → maradoner-0.14.3}/maradoner/main.py

@@ -314,7 +314,7 @@ def _select_motifs(name: str = Argument(..., help='Project name'),
 
 __grn_doc = 'Tests each promoter against each motif per each group. Some people call it GRN.'
 @app.command('grn',
-             help=__select_motif_doc)
+             help=__grn_doc)
 def _grn(name: str = Argument(..., help='Project name'),
          folder: Path = Argument(..., help='Output folder where results will be stored. In total, expect number_of_groups tables of size'
                                            ' comparable to the expression file size.'),
@@ -331,7 +331,7 @@ def _grn(name: str = Argument(..., help='Project name'),
     rprint(f'[green][bold]✔️[/bold] Done![/green]\t time: {dt:.2f} s.')
     
 __estimate_promvar_doc = 'Estimates each promoter variance for each group using empirical Bayesian shrinkage.'\
-                         ' A necessary step before computing GRN.'
+                         ' A very recommended step before computing GRN.'
 @app.command('estimate-promoter-variance',
              help=__estimate_promvar_doc)
 def _estimate_promoter_variance(name: str = Argument(..., help='Project name'),

{maradoner-0.14.1 → maradoner-0.14.3}/maradoner.egg-info/PKG-INFO

@@ -1,17 +1,16 @@
 Metadata-Version: 2.1
 Name: maradoner
-Version: 0.14.1
+Version: 0.14.3
 Summary: Variance-adjusted estimation of motif activities.
-Home-page: https://github.com/autosome-ru/nemara
+Home-page: https://github.com/autosome-ru/maradoner
 Author: Georgy Meshcheryakov
 Author-email: iam@georgy.top
-Classifier: Programming Language :: Python :: 3.9
-Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Scientific/Engineering
 Classifier: Operating System :: OS Independent
-Requires-Python: >=3.9
+Requires-Python: >=3.11
 Description-Content-Type: text/markdown
 Requires-Dist: pip>=24.0
 Requires-Dist: typer>=0.13
@@ -27,6 +26,7 @@ Requires-Dist: dill>=0.3.9
 Requires-Dist: rich>=12.6.0
 Requires-Dist: tqdm>=4.0
 Requires-Dist: scikit-learn>=1.6
+Requires-Dist: tables>=3.10
 
 
 **MARADONER**

{maradoner-0.14.1 → maradoner-0.14.3}/maradoner.egg-info/requires.txt

@@ -12,3 +12,4 @@ dill>=0.3.9
 rich>=12.6.0
 tqdm>=4.0
 scikit-learn>=1.6
+tables>=3.10

{maradoner-0.14.1 → maradoner-0.14.3}/setup.py

@@ -23,13 +23,12 @@ setup(
     author='Georgy Meshcheryakov',
     author_email='iam@georgy.top',
     install_requires=__min_reqs__,
-    python_requires='>=3.9',
-    url="https://github.com/autosome-ru/nemara",
+    python_requires='>=3.11',
+    url="https://github.com/autosome-ru/maradoner",
     classifiers=[
-	      "Programming Language :: Python :: 3.9",
-	      "Programming Language :: Python :: 3.10",
 	      "Programming Language :: Python :: 3.11",
 	      "Programming Language :: Python :: 3.12",
+	      "Programming Language :: Python :: 3.13",
 	      "Topic :: Scientific/Engineering",
               "Operating System :: OS Independent"]