spacr 0.3.2__tar.gz → 0.3.22__tar.gz

{spacr-0.3.2/spacr.egg-info → spacr-0.3.22}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: spacr
-Version: 0.3.2
+Version: 0.3.22
 Summary: Spatial phenotype analysis of crisp screens (SpaCr)
 Home-page: https://github.com/EinarOlafsson/spacr
 Author: Einar Birnir Olafsson
@@ -10,34 +10,41 @@ Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: OS Independent
 Description-Content-Type: text/x-rst
 License-File: LICENSE
-Requires-Dist: torch<3.0,>=2.0
-Requires-Dist: torchvision<1.0,>=0.1
-Requires-Dist: torch-geometric<3.0,>=2.5
 Requires-Dist: numpy<2.0,>=1.26.4
-Requires-Dist: bottleneck<2.0,>=1.3.6
-Requires-Dist: numexpr<3.0,>=2.8.4
 Requires-Dist: pandas<3.0,>=2.2.1
-Requires-Dist: statsmodels<1.0,>=0.14.1
+Requires-Dist: scipy<2.0,>=1.12.0
+Requires-Dist: cellpose<4.0,>=3.0.6
 Requires-Dist: scikit-image<1.0,>=0.22.0
 Requires-Dist: scikit-learn<2.0,>=1.4.1
-Requires-Dist: seaborn<1.0,>=0.13.2
-Requires-Dist: matplotlib<4.0,>=3.8.3
-Requires-Dist: shap<1.0,>=0.45.0
-Requires-Dist: pillow<11.0,>=10.2.0
-Requires-Dist: imageio<3.0,>=2.34.0
-Requires-Dist: scipy<2.0,>=1.12.0
-Requires-Dist: ipywidgets<9.0,>=8.1.2
 Requires-Dist: mahotas<2.0,>=1.4.13
 Requires-Dist: btrack<1.0,>=0.6.5
 Requires-Dist: trackpy<1.0,>=0.6.2
-Requires-Dist: cellpose<4.0,>=3.0.6
-Requires-Dist: IPython<9.0,>=8.18.1
+Requires-Dist: statsmodels<1.0,>=0.14.1
+Requires-Dist: shap<1.0,>=0.45.0
+Requires-Dist: torch<3.0,>=2.0
+Requires-Dist: torchvision<1.0,>=0.1
+Requires-Dist: torch-geometric<3.0,>=2.5
+Requires-Dist: torchcam<1.0,>=0.4.0
+Requires-Dist: transformers<5.0,>=4.45.2
+Requires-Dist: segmentation_models_pytorch>=0.3.3
+Requires-Dist: monai>=1.3.0
+Requires-Dist: captum<1.0,>=0.7.0
+Requires-Dist: seaborn<1.0,>=0.13.2
+Requires-Dist: matplotlib<4.0,>=3.8.3
+Requires-Dist: adjustText<2.0,>=1.2.0
+Requires-Dist: bottleneck<2.0,>=1.3.6
+Requires-Dist: numexpr<3.0,>=2.8.4
 Requires-Dist: opencv-python-headless<5.0,>=4.9.0.80
+Requires-Dist: pillow<11.0,>=10.2.0
+Requires-Dist: tifffile>=2023.4.12
+Requires-Dist: nd2reader<4.0,>=3.3.0
+Requires-Dist: czifile
+Requires-Dist: imageio<3.0,>=2.34.0
+Requires-Dist: pingouin<1.0,>=0.5.5
 Requires-Dist: umap-learn<1.0,>=0.5.6
 Requires-Dist: ttkthemes<4.0,>=3.2.2
 Requires-Dist: xgboost<3.0,>=2.0.3
 Requires-Dist: PyWavelets<2.0,>=1.6.0
-Requires-Dist: torchcam<1.0,>=0.4.0
 Requires-Dist: ttf_opensans>=2020.10.30
 Requires-Dist: customtkinter<6.0,>=5.2.2
 Requires-Dist: biopython<2.0,>=1.80
@@ -50,19 +57,15 @@ Requires-Dist: tables<4.0,>=3.8.0
 Requires-Dist: rapidfuzz<4.0,>=3.9
 Requires-Dist: keyring<16.0,>=15.1
 Requires-Dist: screeninfo<1.0,>=0.8.1
-Requires-Dist: ipykernel
-Requires-Dist: gdown
 Requires-Dist: fastremap>=1.14.1
-Requires-Dist: monai>=1.3.0
 Requires-Dist: pytz>=2023.3.post1
-Requires-Dist: segmentation_models_pytorch>=0.3.3
-Requires-Dist: tifffile>=2023.4.12
 Requires-Dist: tqdm>=4.65.0
 Requires-Dist: wandb>=0.16.2
 Requires-Dist: openai<2.0,>=1.50.2
-Requires-Dist: nd2reader<4.0,>=3.3.0
-Requires-Dist: czifile
-Requires-Dist: adjustText<2.0,>=1.2.0
+Requires-Dist: gdown
+Requires-Dist: IPython<9.0,>=8.18.1
+Requires-Dist: ipykernel
+Requires-Dist: ipywidgets<9.0,>=8.1.2
 Requires-Dist: huggingface-hub<0.25,>=0.24.0
 Provides-Extra: dev
 Requires-Dist: pytest<3.11,>=3.9; extra == "dev"

spacr-0.3.22/notebooks/cv_scoring_nb.ipynb

@@ -0,0 +1,171 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b370394d",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [],
+   "source": [
+    "# Description: Generate train and test folders with class subfolders for training DL models.\n",
+    "from spacr.io import generate_training_dataset\n",
+    "\n",
+    "settings = {'src':'path or list of paths',\n",
+    "            'dataset_mode':'metadata',\n",
+    "            'test_split':0.1,\n",
+    "            'metadata_type_by':'col',\n",
+    "            'class_metadata':[['c1'],['c2']],\n",
+    "            'png_type':'cell_png',\n",
+    "            'nuclei_limit':True,\n",
+    "            'pathogen_limit':3,\n",
+    "            'uninfected':False,\n",
+    "            'size':None}\n",
+    "\n",
+    "generate_training_dataset(settings)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "cb293108",
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [],
+   "source": [
+    "# Description: Train a torch model to classify single object images\n",
+    "from spacr.deep_spacr import train_test_model\n",
+    "\n",
+    "settings = {'src':'path', \n",
+    "            'train':True,\n",
+    "            'test': False,\n",
+    "            'custom_model':False,\n",
+    "            'custom_model_path':None,\n",
+    "            'classes':['nc','pc'],\n",
+    "            'model_type':'maxvit_t',\n",
+    "            'optimizer_type':'adamw',\n",
+    "            'schedule':'reduce_lr_on_plateau', #reduce_lr_on_plateau, step_lr\n",
+    "            'loss_type':'focal_loss', #binary_cross_entropy_with_logits, #focal_loss\n",
+    "            'normalize':True,\n",
+    "            'image_size':224,\n",
+    "            'batch_size':64,\n",
+    "            'epochs':100,\n",
+    "            'val_split':0.1,\n",
+    "            'learning_rate':0.0001,\n",
+    "            'weight_decay':0.00001,\n",
+    "            'dropout_rate':0.1,\n",
+    "            'init_weights':True,\n",
+    "            'amsgrad':True,\n",
+    "            'use_checkpoint':True,\n",
+    "            'gradient_accumulation':True,\n",
+    "            'gradient_accumulation_steps':4,\n",
+    "            'intermedeate_save':True,\n",
+    "            'pin_memory':True,\n",
+    "            'n_jobs':30,\n",
+    "            'train_channels':['r','g','b'],\n",
+    "            'augment':False,\n",
+    "            'verbose':True}\n",
+    "\n",
+    "train_test_model(settings)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "09107d9a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Description: Generate a tar file containing single object images.\n",
+    "from spacr.io import generate_dataset\n",
+    "\n",
+    "settings = {'src':'path or list of paths',\n",
+    "           'file_metadata':None,\n",
+    "           'experiment':'tsg101_screen_plate1',\n",
+    "           'sample':None}\n",
+    "\n",
+    "generate_dataset(settings)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c52af6e8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Description: Classify images in a tar dataset with a trained torch model.\n",
+    "from spacr.core import apply_model_to_tar\n",
+    "\n",
+    "settings = {'tar_path':'path',\n",
+    "            'model_path':'path', \n",
+    "            'file_type':'cell_png',\n",
+    "            'image_size':224,\n",
+    "            'batch_size':64,\n",
+    "            'normalize':True,\n",
+    "            'score_threshold':0.5,\n",
+    "            'n_jobs':30,\n",
+    "            'verbose':True}\n",
+    "\n",
+    "result_df = spacr.core.apply_model_to_tar(settings)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "5b36a8ec",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Description: Fix a regression model to estimate the effect size of gRNAs on cell scores.\n",
+    "# \n",
+    "\n",
+    "from spacr.ml import perform_regression\n",
+    "import pandas as pd\n",
+    "%matplotlib inline\n",
+    "\n",
+    "settings = {'count_data':'path',\n",
+    "            'score_data':'path',\n",
+    "            'highlight':'string',\n",
+    "            'fraction_threshold':0.1,\n",
+    "            'dependent_variable': 'prediction_probability_class_1',\n",
+    "            'transform':'log',\n",
+    "            'agg_type':'median',\n",
+    "            'min_cell_count':25,\n",
+    "            'regression_type':'ols',\n",
+    "            'random_row_column_effects':False,\n",
+    "            'plate':None,\n",
+    "            'cov_type':None,\n",
+    "            'alpha':0.8,\n",
+    "            'nc':'c1',\n",
+    "            'pc':'c2',\n",
+    "            'other':'c3'}\n",
+    "\n",
+    "coef_df = perform_regression(settings)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "spacr",
+   "language": "python",
+   "name": "spacr"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.19"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

spacr-0.3.22/notebooks/machine_learning_spacr_nb.ipynb

@@ -0,0 +1,101 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "9e030e83",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [],
+   "source": [
+    "# Description: Train a ML model to classigy cells based on measurement data\n",
+    "\n",
+    "from spacr.ml import generate_ml_scores\n",
+    "%matplotlib inline\n",
+    "\n",
+    "settings = {'src':'path',\n",
+    "            'model_type_ml':'xgboost',\n",
+    "            'heatmap_feature':'predictions',\n",
+    "            'grouping':'mean',\n",
+    "            'min_max':'allq',\n",
+    "            'cmap':'viridis',\n",
+    "            'n_estimators':100,\n",
+    "            'test_size':0.2,\n",
+    "            'location_column':'col',\n",
+    "            'positive_control':'c2',\n",
+    "            'negative_control':'c1',\n",
+    "            'exclude':None,\n",
+    "            'include_multinucleated':True,\n",
+    "            'include_multiinfected':3,\n",
+    "            'include_noninfected':True,\n",
+    "            'n_repeats':10,\n",
+    "            'top_features':30,\n",
+    "            'channel_of_interest':1,\n",
+    "            'minimum_cell_count':25,\n",
+    "            'remove_low_variance_features':True,\n",
+    "            'remove_highly_correlated_features':True,\n",
+    "            'verbose':False,\n",
+    "            'n_jobs':10}\n",
+    "\n",
+    "results = spacr.ml.generate_ml_scores(settings)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "e316cb19",
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [],
+   "source": [
+    "# Description: Fix a regression model to estimate the effect size of gRNAs on cell scores.\n",
+    "\n",
+    "from spacr.ml import perform_regression\n",
+    "import pandas as pd\n",
+    "%matplotlib inline\n",
+    "\n",
+    "settings = {'count_data':'path',\n",
+    "            'score_data':'path',\n",
+    "            'highlight':'string',\n",
+    "            'fraction_threshold':0.1,\n",
+    "            'dependent_variable': 'prediction_probability_class_1',\n",
+    "            'transform':'log',\n",
+    "            'agg_type':'median',\n",
+    "            'min_cell_count':25,\n",
+    "            'regression_type':'ols',\n",
+    "            'random_row_column_effects':False,\n",
+    "            'plate':None,\n",
+    "            'cov_type':None,\n",
+    "            'alpha':0.8,\n",
+    "            'nc':'c1',\n",
+    "            'pc':'c2',\n",
+    "            'other':'c3'}\n",
+    "\n",
+    "coef_df = perform_regression(settings)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "spacr",
+   "language": "python",
+   "name": "spacr"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.19"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

{spacr-0.3.2 → spacr-0.3.22}/setup.py

@@ -6,34 +6,41 @@ with open("README.rst", "r", encoding="utf-8") as fh:
     long_description = fh.read()
 
 dependencies = [
-    'torch>=2.0,<3.0',
-    'torchvision>=0.1,<1.0',
-    'torch-geometric>=2.5,<3.0',
     'numpy>=1.26.4,<2.0',
-    'bottleneck>=1.3.6,<2.0',
-    'numexpr>=2.8.4,<3.0',
     'pandas>=2.2.1,<3.0',
-    'statsmodels>=0.14.1,<1.0',
+    'scipy>=1.12.0,<2.0',
+    'cellpose>=3.0.6,<4.0',
     'scikit-image>=0.22.0,<1.0',
     'scikit-learn>=1.4.1,<2.0',
-    'seaborn>=0.13.2,<1.0',
-    'matplotlib>=3.8.3,<4.0',
-    'shap>=0.45.0,<1.0',
-    'pillow>=10.2.0,<11.0',
-    'imageio>=2.34.0,<3.0',
-    'scipy>=1.12.0,<2.0',
-    'ipywidgets>=8.1.2,<9.0',
     'mahotas>=1.4.13,<2.0',
     'btrack>=0.6.5,<1.0',
     'trackpy>=0.6.2,<1.0',
-    'cellpose>=3.0.6,<4.0',
-    'IPython>=8.18.1,<9.0',
+    'statsmodels>=0.14.1,<1.0',
+    'shap>=0.45.0,<1.0',
+    'torch>=2.0,<3.0',
+    'torchvision>=0.1,<1.0',
+    'torch-geometric>=2.5,<3.0',
+    'torchcam>=0.4.0,<1.0',
+    'transformers>=4.45.2, <5.0',
+    'segmentation_models_pytorch>=0.3.3',
+    'monai>=1.3.0',
+    'captum>=0.7.0, <1.0',
+    'seaborn>=0.13.2,<1.0',
+    'matplotlib>=3.8.3,<4.0',
+    'adjustText>=1.2.0,<2.0',
+    'bottleneck>=1.3.6,<2.0',
+    'numexpr>=2.8.4,<3.0',
     'opencv-python-headless>=4.9.0.80,<5.0',
+    'pillow>=10.2.0,<11.0',
+    'tifffile>=2023.4.12',
+    'nd2reader>=3.3.0, <4.0',
+    'czifile',
+    'imageio>=2.34.0,<3.0',
+    'pingouin>=0.5.5,<1.0',
     'umap-learn>=0.5.6,<1.0',
     'ttkthemes>=3.2.2,<4.0',
     'xgboost>=2.0.3,<3.0',
     'PyWavelets>=1.6.0,<2.0',
-    'torchcam>=0.4.0,<1.0',
     'ttf_opensans>=2020.10.30',
     'customtkinter>=5.2.2,<6.0', 
     'biopython>=1.80,<2.0',
@@ -46,25 +53,21 @@ dependencies = [
     'rapidfuzz>=3.9, <4.0',
     'keyring>=15.1, <16.0',
     'screeninfo>=0.8.1,<1.0',
-    'ipykernel',
-    'gdown',
     'fastremap>=1.14.1',
-    'monai>=1.3.0',
     'pytz>=2023.3.post1',
-    'segmentation_models_pytorch>=0.3.3',
-    'tifffile>=2023.4.12',
     'tqdm>=4.65.0',
     'wandb>=0.16.2',
     'openai>=1.50.2, <2.0',
-    'nd2reader>=3.3.0, <4.0',
-    'czifile',
-    'adjustText>=1.2.0,<2.0',
+    'gdown',
+    'IPython>=8.18.1,<9.0',
+    'ipykernel',
+    'ipywidgets>=8.1.2,<9.0',
     'huggingface-hub>=0.24.0,<0.25'
 ]
 
 setup(
     name="spacr",
-    version="0.3.2",
+    version="0.3.22",
     author="Einar Birnir Olafsson",
     author_email="olafsson@med.umich.com",
     description="Spatial phenotype analysis of crisp screens (SpaCr)",
@@ -114,4 +117,3 @@ for dep in deps:
         subprocess.run(['pip', 'install', dep], check=True)
     except subprocess.CalledProcessError:
         pass
-

{spacr-0.3.2 → spacr-0.3.22}/spacr/core.py

@@ -844,4 +844,108 @@ def generate_mediar_masks(src, settings, object_type):
         gc.collect()
         torch.cuda.empty_cache()
 
-    print("Mask generation completed.")
+    print("Mask generation completed.")
+
+def generate_screen_graphs(settings):
+    """
+    Generate screen graphs for different measurements in a given source directory.
+
+    Args:
+        src (str or list): Path(s) to the source directory or directories.
+        tables (list): List of tables to include in the analysis (default: ['cell', 'nucleus', 'pathogen', 'cytoplasm']).
+        graph_type (str): Type of graph to generate (default: 'bar').
+        summary_func (str or function): Function to summarize data (default: 'mean').
+        y_axis_start (float): Starting value for the y-axis (default: 0).
+        error_bar_type (str): Type of error bar to use ('std' or 'sem') (default: 'std').
+        theme (str): Theme for the graph (default: 'pastel').
+        representation (str): Representation for grouping (default: 'well').
+        
+    Returns:
+        figs (list): List of generated figures.
+        results (list): List of corresponding result DataFrames.
+    """
+    
+    from .plot import spacrGraph
+    from .io import _read_and_merge_data
+    from.utils import annotate_conditions
+
+    if isinstance(settings['src'], str):
+        srcs = [settings['src']]
+    else:
+        srcs = settings['src']
+
+    all_df = pd.DataFrame()
+    figs = []
+    results = []
+
+    for src in srcs:
+        db_loc = [os.path.join(src, 'measurements', 'measurements.db')]
+        
+        # Read and merge data from the database
+        df, _ = _read_and_merge_data(db_loc, settings['tables'], verbose=True, nuclei_limit=settings['nuclei_limit'], pathogen_limit=settings['pathogen_limit'], uninfected=settings['uninfected'])
+        
+        # Annotate the data
+        df = annotate_conditions(df, cells=settings['cells'], cell_loc=None, pathogens=settings['controls'], pathogen_loc=settings['controls_loc'], treatments=None, treatment_loc=None)
+        
+        # Calculate recruitment metric
+        df['recruitment'] = df['pathogen_channel_1_mean_intensity'] / df['cytoplasm_channel_1_mean_intensity']
+                
+        # Combine with the overall DataFrame
+        all_df = pd.concat([all_df, df], ignore_index=True)
+    
+        # Generate individual plot
+        plotter = spacrGraph(df,
+                             grouping_column='pathogen',
+                             data_column='recruitment',
+                             graph_type=settings['graph_type'],
+                             summary_func=settings['summary_func'],
+                             y_axis_start=settings['y_axis_start'],
+                             error_bar_type=settings['error_bar_type'],
+                             theme=settings['theme'],
+                             representation=settings['representation'])
+
+        plotter.create_plot()
+        fig = plotter.get_figure()
+        results_df = plotter.get_results()
+        
+        # Append to the lists
+        figs.append(fig)
+        results.append(results_df)
+    
+    # Generate plot for the combined data (all_df)
+    plotter = spacrGraph(all_df,
+                         grouping_column='pathogen',
+                         data_column='recruitment',
+                         graph_type=settings['graph_type'],
+                         summary_func=settings['summary_func'],
+                         y_axis_start=settings['y_axis_start'],
+                         error_bar_type=settings['error_bar_type'],
+                         theme=settings['theme'],
+                         representation=settings['representation'])
+
+    plotter.create_plot()
+    fig = plotter.get_figure()
+    results_df = plotter.get_results()
+    
+    figs.append(fig)
+    results.append(results_df)
+    
+    # Save figures and results
+    for i, fig in enumerate(figs):
+        res = results[i]
+        
+        if i < len(srcs):
+            source = srcs[i]
+        else:
+            source = srcs[0]
+
+        # Ensure the destination folder exists
+        dst = os.path.join(source, 'results')
+        print(f"Savings results to {dst}")
+        os.makedirs(dst, exist_ok=True)
+        
+        # Save the figure and results DataFrame
+        fig.savefig(os.path.join(dst, f"figure_controls_{i}_{settings['representation']}_{settings['summary_func']}_{settings['graph_type']}.pdf"), format='pdf')
+        res.to_csv(os.path.join(dst, f"results_controls_{i}_{settings['representation']}_{settings['summary_func']}_{settings['graph_type']}.csv"), index=False)
+
+    return