spacr 0.0.70__tar.gz → 0.0.80__tar.gz

{spacr-0.0.70/spacr.egg-info → spacr-0.0.80}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: spacr
-Version: 0.0.70
+Version: 0.0.80
 Summary: Spatial phenotype analysis of crisp screens (SpaCr)
 Home-page: https://github.com/EinarOlafsson/spacr
 Author: Einar Birnir Olafsson
@@ -9,7 +9,7 @@ Classifier: Programming Language :: Python :: 3
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: OS Independent
 License-File: LICENSE
-Requires-Dist: dgl
+Requires-Dist: dgl==0.9.1
 Requires-Dist: torch<3.0,>=2.2.1
 Requires-Dist: torchvision<1.0,>=0.17.1
 Requires-Dist: torch-geometric<3.0,>=2.5.1
@@ -46,6 +46,8 @@ Requires-Dist: opencv-python-headless; extra == "headless"
 Provides-Extra: full
 Requires-Dist: opencv-python; extra == "full"
 
+.. |Documentation Status| image:: https://readthedocs.org/projects/spacr/badge/?version=latest
+   :target: https://spacr.readthedocs.io/en/latest/?badge=latest
 .. |PyPI version| image:: https://badge.fury.io/py/spacr.svg
    :target: https://badge.fury.io/py/spacr
 .. |Python version| image:: https://img.shields.io/pypi/pyversions/spacr
@@ -55,25 +57,25 @@ Requires-Dist: opencv-python; extra == "full"
 .. |repo size| image:: https://img.shields.io/github/repo-size/EinarOlafsson/spacr
    :target: https://github.com/EinarOlafsson/spacr/
 
-|PyPI version| |Python version| |Licence: GPL v3| |repo size|
+|Documentation Status| |PyPI version| |Python version| |Licence: GPL v3| |repo size|
 
 SpaCr
 =====
 
-Spatial phenotype analysis of CRISPR-Cas9 screens (SpaCr). The spatial organization of organelles and proteins within cells constitutes a key level of functional regulation. In the context of infectious disease, the spatial relationships between host cell structures and intracellular pathogens are critical to understand host clearance mechanisms and how pathogens evade them. Spacr is a Python-based software package for generating single cell image data for deep-learning sub-cellular/cellular phenotypic classification from pooled genetic CRISPR-Cas9 screens. Spacr provides a flexible toolset to extract single cell images and measurements from high content cell painting experiments, train deep-learning models to classify cellular/subcellular phenotypes, simulate and analyze pooled CRISPR-Cas9 imaging screens.
+Spatial phenotype analysis of CRISPR-Cas9 screens (SpaCr). The spatial organization of organelles and proteins within cells constitutes a key level of functional regulation. In the context of infectious disease, the spatial relationships between host cell structures and intracellular pathogens are critical to understand host clearance mechanisms and how pathogens evade them. SpaCr is a Python-based software package for generating single-cell image data for deep-learning sub-cellular/cellular phenotypic classification from pooled genetic CRISPR-Cas9 screens. SpaCr provides a flexible toolset to extract single-cell images and measurements from high-content cell painting experiments, train deep-learning models to classify cellular/subcellular phenotypes, simulate, and analyze pooled CRISPR-Cas9 imaging screens.
 
 Features
 --------
 
 -  **Generate Masks:** Generate cellpose masks of cell, nuclei, and pathogen objects.
 
--  **Object Measurements:** Measurements for each object including scikit-image-regionprops, intensity percentiles, shannon-entropy, pearsons and manders correlations, homogeneity and radial distribution. Measurements are saved to a SQL database in object level tables.
+-  **Object Measurements:** Measurements for each object including scikit-image-regionprops, intensity percentiles, shannon-entropy, pearsons and manders correlations, homogeneity, and radial distribution. Measurements are saved to a SQL database in object-level tables.
 
--  **Crop Images:** Objects (e.g. cells) can be saved as PNGs from the object area or bounding box area of each object. Object paths are saved in a SQL database that can be annotated and used to train CNNs/Transformer models for classification tasks.
+-  **Crop Images:** Objects (e.g., cells) can be saved as PNGs from the object area or bounding box area of each object. Object paths are saved in a SQL database that can be annotated and used to train CNNs/Transformer models for classification tasks.
 
 -  **Train CNNs or Transformers:** Train Torch Convolutional Neural Networks (CNNs) or Transformers to classify single object images. Train Torch models with IRM/ERM, checkpointing.
 
--  **Manual Annotation:** Supports manual annotation of single cell images and segmentation to refine training datasets for training CNNs/Transformers or cellpose, respectively.
+-  **Manual Annotation:** Supports manual annotation of single-cell images and segmentation to refine training datasets for training CNNs/Transformers or cellpose, respectively.
 
 -  **Finetune Cellpose Models:** Adjust pre-existing Cellpose models to your specific dataset for improved performance.
 
@@ -93,7 +95,7 @@ Requires Tkinter for graphical user interface features.
 Ubuntu
 ~~~~~~
 
-Before installing spacr, ensure Tkinter is installed:
+Before installing SpaCr, ensure Tkinter is installed:
 
 (Tkinter is included with the standard Python installation on macOS, and Windows)
 

{spacr-0.0.70 → spacr-0.0.80}/README.rst

@@ -1,3 +1,5 @@
+.. |Documentation Status| image:: https://readthedocs.org/projects/spacr/badge/?version=latest
+   :target: https://spacr.readthedocs.io/en/latest/?badge=latest
 .. |PyPI version| image:: https://badge.fury.io/py/spacr.svg
    :target: https://badge.fury.io/py/spacr
 .. |Python version| image:: https://img.shields.io/pypi/pyversions/spacr
@@ -7,25 +9,25 @@
 .. |repo size| image:: https://img.shields.io/github/repo-size/EinarOlafsson/spacr
    :target: https://github.com/EinarOlafsson/spacr/
 
-|PyPI version| |Python version| |Licence: GPL v3| |repo size|
+|Documentation Status| |PyPI version| |Python version| |Licence: GPL v3| |repo size|
 
 SpaCr
 =====
 
-Spatial phenotype analysis of CRISPR-Cas9 screens (SpaCr). The spatial organization of organelles and proteins within cells constitutes a key level of functional regulation. In the context of infectious disease, the spatial relationships between host cell structures and intracellular pathogens are critical to understand host clearance mechanisms and how pathogens evade them. Spacr is a Python-based software package for generating single cell image data for deep-learning sub-cellular/cellular phenotypic classification from pooled genetic CRISPR-Cas9 screens. Spacr provides a flexible toolset to extract single cell images and measurements from high content cell painting experiments, train deep-learning models to classify cellular/subcellular phenotypes, simulate and analyze pooled CRISPR-Cas9 imaging screens.
+Spatial phenotype analysis of CRISPR-Cas9 screens (SpaCr). The spatial organization of organelles and proteins within cells constitutes a key level of functional regulation. In the context of infectious disease, the spatial relationships between host cell structures and intracellular pathogens are critical to understand host clearance mechanisms and how pathogens evade them. SpaCr is a Python-based software package for generating single-cell image data for deep-learning sub-cellular/cellular phenotypic classification from pooled genetic CRISPR-Cas9 screens. SpaCr provides a flexible toolset to extract single-cell images and measurements from high-content cell painting experiments, train deep-learning models to classify cellular/subcellular phenotypes, simulate, and analyze pooled CRISPR-Cas9 imaging screens.
 
 Features
 --------
 
 -  **Generate Masks:** Generate cellpose masks of cell, nuclei, and pathogen objects.
 
--  **Object Measurements:** Measurements for each object including scikit-image-regionprops, intensity percentiles, shannon-entropy, pearsons and manders correlations, homogeneity and radial distribution. Measurements are saved to a SQL database in object level tables.
+-  **Object Measurements:** Measurements for each object including scikit-image-regionprops, intensity percentiles, shannon-entropy, pearsons and manders correlations, homogeneity, and radial distribution. Measurements are saved to a SQL database in object-level tables.
 
--  **Crop Images:** Objects (e.g. cells) can be saved as PNGs from the object area or bounding box area of each object. Object paths are saved in a SQL database that can be annotated and used to train CNNs/Transformer models for classification tasks.
+-  **Crop Images:** Objects (e.g., cells) can be saved as PNGs from the object area or bounding box area of each object. Object paths are saved in a SQL database that can be annotated and used to train CNNs/Transformer models for classification tasks.
 
 -  **Train CNNs or Transformers:** Train Torch Convolutional Neural Networks (CNNs) or Transformers to classify single object images. Train Torch models with IRM/ERM, checkpointing.
 
--  **Manual Annotation:** Supports manual annotation of single cell images and segmentation to refine training datasets for training CNNs/Transformers or cellpose, respectively.
+-  **Manual Annotation:** Supports manual annotation of single-cell images and segmentation to refine training datasets for training CNNs/Transformers or cellpose, respectively.
 
 -  **Finetune Cellpose Models:** Adjust pre-existing Cellpose models to your specific dataset for improved performance.
 
@@ -45,7 +47,7 @@ Requires Tkinter for graphical user interface features.
 Ubuntu
 ~~~~~~
 
-Before installing spacr, ensure Tkinter is installed:
+Before installing SpaCr, ensure Tkinter is installed:
 
 (Tkinter is included with the standard Python installation on macOS, and Windows)
 

{spacr-0.0.70 → spacr-0.0.80}/setup.py

@@ -13,13 +13,9 @@ def get_cuda_version():
 cuda_version = get_cuda_version()
 
 if cuda_version:
-    dgl_dependency = f'dgl-cu{cuda_version}'
+    dgl_dependency = f'dgl-cu{cuda_version}==0.9.1'  # Specify the version of DGL compatible with your setup
 else:
-    dgl_dependency = 'dgl'  # Fallback to CPU version if no CUDA is detected
-
-# Ensure you have read the README.rst content into a variable, e.g., `long_description`
-#with open("README.md", "r", encoding="utf-8") as fh:
-#    long_description = fh.read()
+    dgl_dependency = 'dgl==0.9.1'  # Fallback to CPU version if no CUDA is detected
 
 # Ensure you have read the README.rst content into a variable, e.g., `long_description`
 with open("README.rst", "r", encoding="utf-8") as fh:
@@ -60,7 +56,7 @@ dependencies = [
 
 setup(
     name="spacr",
-    version="0.0.70",
+    version="0.0.80",
     author="Einar Birnir Olafsson",
     author_email="olafsson@med.umich.com",
     description="Spatial phenotype analysis of crisp screens (SpaCr)",

{spacr-0.0.70 → spacr-0.0.80}/spacr/__init__.py

@@ -8,16 +8,18 @@ from . import utils
 from . import plot
 from . import measure
 from . import sim
+from . import sequencing
 from . import timelapse
 from . import deep_spacr
-from . import mask_app
 from . import annotate_app
 from . import gui_utils
+from . import mask_app
 from . import gui_mask_app
 from . import gui_measure_app
 from . import gui_classify_app
 from . import logger
 
+
 __all__ = [
     "core",
     "io",
@@ -25,6 +27,7 @@ __all__ = [
     "plot",
     "measure",
     "sim",
+    "sequencing"
     "timelapse",
     "deep_spacr",
     "annotate_app",

spacr-0.0.80/spacr/__main__.py

@@ -0,0 +1,6 @@
+"""
+Copyright © 2024 Something
+"""
+
+if __name__ == "__main__":
+    main()

{spacr-0.0.70 → spacr-0.0.80}/spacr/annotate_app.py

@@ -10,13 +10,16 @@ from IPython.display import display, HTML
 import tkinter as tk
 from tkinter import ttk
 from ttkthemes import ThemedTk
+from skimage.exposure import rescale_intensity
+import cv2
+import matplotlib.pyplot as plt
 
 from .logger import log_function_call
 
 from .gui_utils import ScrollableFrame, set_default_font, set_dark_style, create_dark_mode, style_text_boxes, create_menu_bar
 
 class ImageApp:
-    def __init__(self, root, db_path, src, image_type=None, channels=None, grid_rows=None, grid_cols=None, image_size=(200, 200), annotation_column='annotate'):
+    def __init__(self, root, db_path, src, image_type=None, channels=None, grid_rows=None, grid_cols=None, image_size=(200, 200), annotation_column='annotate', normalize=False, percentiles=(1,99)):
         """
         Initializes an instance of the ImageApp class.
 
@@ -30,6 +33,7 @@ class ImageApp:
         - grid_cols (int): The number of columns in the image grid.
         - image_size (tuple): The size of the displayed images.
         - annotation_column (str): The column name for image annotations in the database.
+        - normalize (bool): Whether to normalize images to their 2nd and 98th percentiles. Defaults to False.
         """
         self.root = root
         self.db_path = db_path
@@ -41,6 +45,8 @@ class ImageApp:
         self.annotation_column = annotation_column
         self.image_type = image_type
         self.channels = channels
+        self.normalize = normalize
+        self.percentiles = percentiles
         self.images = {}
         self.pending_updates = {}
         self.labels = []
@@ -119,49 +125,80 @@ class ImageApp:
             label.bind('<Button-3>', self.get_on_image_click(path, label, img))
 
         self.root.update()
-        
+
+    def load_single_image(self, path_annotation_tuple):
+        """
+        Loads a single image from the given path and annotation tuple.
+
+        Args:
+            path_annotation_tuple (tuple): A tuple containing the image path and its annotation.
+
+        Returns:
+            img (PIL.Image.Image): The loaded image.
+            annotation: The annotation associated with the image.
+        """
+        path, annotation = path_annotation_tuple
+        img = Image.open(path)
+        img = self.normalize_image(img, self.normalize, self.percentiles)
+        img = img.convert('RGB')
+        img = self.filter_channels(img)
+        img = img.resize(self.image_size)
+        return img, annotation
+
     @staticmethod
-    def normalize_image(img):
+    def normalize_image(img, normalize=False, percentiles=(1, 99)):
         """
-        Normalize the pixel values of an image to the range [0, 255].
+        Normalize the pixel values of an image based on the 2nd and 98th percentiles or the image min and max values,
+        and ensure the image is exported as 8-bit.
 
         Parameters:
-        - img: PIL.Image.Image
-            The input image to be normalized.
+        - img: PIL.Image.Image. The input image to be normalized.
+        - normalize: bool. Whether to normalize based on the 2nd and 98th percentiles.
+        - percentiles: tuple. The percentiles to use for normalization.
 
         Returns:
-        - PIL.Image.Image
-            The normalized image.
+        - PIL.Image.Image. The normalized and 8-bit converted image.
         """
         img_array = np.array(img)
-        img_array = ((img_array - img_array.min()) * (1/(img_array.max() - img_array.min()) * 255)).astype('uint8')
-        return Image.fromarray(img_array)
 
+        if normalize:
+            if img_array.ndim == 2:  # Grayscale image
+                p2, p98 = np.percentile(img_array, percentiles)
+                img_array = rescale_intensity(img_array, in_range=(p2, p98), out_range=(0, 255))
+            else:  # Color image or multi-channel image
+                for channel in range(img_array.shape[2]):
+                    p2, p98 = np.percentile(img_array[:, :, channel], percentiles)
+                    img_array[:, :, channel] = rescale_intensity(img_array[:, :, channel], in_range=(p2, p98), out_range=(0, 255))
+
+        img_array = np.clip(img_array, 0, 255).astype('uint8')
+
+        return Image.fromarray(img_array)
+    
     def add_colored_border(self, img, border_width, border_color):
-            """
-            Adds a colored border to an image.
-
-            Args:
-                img (PIL.Image.Image): The input image.
-                border_width (int): The width of the border in pixels.
-                border_color (str): The color of the border in RGB format.
-
-            Returns:
-                PIL.Image.Image: The image with the colored border.
-            """
-            top_border = Image.new('RGB', (img.width, border_width), color=border_color)
-            bottom_border = Image.new('RGB', (img.width, border_width), color=border_color)
-            left_border = Image.new('RGB', (border_width, img.height), color=border_color)
-            right_border = Image.new('RGB', (border_width, img.height), color=border_color)
-
-            bordered_img = Image.new('RGB', (img.width + 2 * border_width, img.height + 2 * border_width), color='white')
-            bordered_img.paste(top_border, (border_width, 0))
-            bordered_img.paste(bottom_border, (border_width, img.height + border_width))
-            bordered_img.paste(left_border, (0, border_width))
-            bordered_img.paste(right_border, (img.width + border_width, border_width))
-            bordered_img.paste(img, (border_width, border_width))
-
-            return bordered_img
+        """
+        Adds a colored border to an image.
+
+        Args:
+            img (PIL.Image.Image): The input image.
+            border_width (int): The width of the border in pixels.
+            border_color (str): The color of the border in RGB format.
+
+        Returns:
+            PIL.Image.Image: The image with the colored border.
+        """
+        top_border = Image.new('RGB', (img.width, border_width), color=border_color)
+        bottom_border = Image.new('RGB', (img.width, border_width), color=border_color)
+        left_border = Image.new('RGB', (border_width, img.height), color=border_color)
+        right_border = Image.new('RGB', (border_width, img.height), color=border_color)
+
+        bordered_img = Image.new('RGB', (img.width + 2 * border_width, img.height + 2 * border_width), color='white')
+        bordered_img.paste(top_border, (border_width, 0))
+        bordered_img.paste(bottom_border, (border_width, img.height + border_width))
+        bordered_img.paste(left_border, (0, border_width))
+        bordered_img.paste(right_border, (img.width + border_width, border_width))
+        bordered_img.paste(img, (border_width, border_width))
+
+        return bordered_img
     
     def filter_channels(self, img):
         """
@@ -189,26 +226,6 @@ class ImageApp:
 
         return Image.merge("RGB", (r, g, b))
 
-    def load_single_image(self, path_annotation_tuple):
-            """
-            Loads a single image from the given path and annotation tuple.
-
-            Args:
-                path_annotation_tuple (tuple): A tuple containing the image path and its annotation.
-
-            Returns:
-                img (PIL.Image.Image): The loaded image.
-                annotation: The annotation associated with the image.
-            """
-            path, annotation = path_annotation_tuple
-            img = Image.open(path)
-            if img.mode == "I":
-                img = self.normalize_image(img)
-            img = img.convert('RGB')
-            img = self.filter_channels(img)
-            img = img.resize(self.image_size)
-            return img, annotation
-        
     def get_on_image_click(self, path, label, img):
         """
         Returns a callback function that handles the click event on an image.
@@ -244,7 +261,7 @@ class ImageApp:
             self.root.update()
 
         return on_image_click
-     
+
     @staticmethod
     def update_html(text):
         display(HTML(f"""
@@ -349,7 +366,7 @@ class ImageApp:
         self.root.destroy()
         print(f'Quit application')
 
-def annotate(src, image_type=None, channels=None, geom="1000x1100", img_size=(200, 200), rows=5, columns=5, annotation_column='annotate'):
+def annotate(src, image_type=None, channels=None, geom="1000x1100", img_size=(200, 200), rows=5, columns=5, annotation_column='annotate', normalize=False, percentiles=(1,99)):
     """
     Annotates images in a database using a graphical user interface.
 
@@ -363,11 +380,9 @@ def annotate(src, image_type=None, channels=None, geom="1000x1100", img_size=(20
         rows (int, optional): The number of rows in the image grid. Defaults to 5.
         columns (int, optional): The number of columns in the image grid. Defaults to 5.
         annotation_column (str, optional): The name of the annotation column in the database table. Defaults to 'annotate'.
+        normalize (bool, optional): Whether to normalize images to their 2nd and 98th percentiles. Defaults to False.
     """
     db = os.path.join(src, 'measurements/measurements.db')
-    #print('src', src)
-    #print('db', db)
-
     conn = sqlite3.connect(db)
     c = conn.cursor()
     c.execute('PRAGMA table_info(png_list)')
@@ -379,7 +394,7 @@ def annotate(src, image_type=None, channels=None, geom="1000x1100", img_size=(20
 
     root = tk.Tk()
     root.geometry(geom)
-    app = ImageApp(root, db, src, image_type=image_type, channels=channels, image_size=img_size, grid_rows=rows, grid_cols=columns, annotation_column=annotation_column)
+    app = ImageApp(root, db, src, image_type=image_type, channels=channels, image_size=img_size, grid_rows=rows, grid_cols=columns, annotation_column=annotation_column, normalize=normalize, percentiles=percentiles)
     next_button = tk.Button(root, text="Next", command=app.next_page)
     next_button.grid(row=app.grid_rows, column=app.grid_cols - 1)
     back_button = tk.Button(root, text="Back", command=app.previous_page)
@@ -390,7 +405,6 @@ def annotate(src, image_type=None, channels=None, geom="1000x1100", img_size=(20
     app.load_images()
     root.mainloop()
 
-
 def check_for_duplicates(db):
     """
     Check for duplicates in the given SQLite database.