Merge branch 'main' into win-ci

Author: constantinpape

.github/workflows/build_docs.yaml

@@ -25,12 +25,11 @@ jobs:
     - name: Set up Micromamba
       uses: mamba-org/setup-micromamba@v2
       with:
-        micromamba-version: "latest"       # Use the latest version of micromamba
-        environment-file: environment_cpu.yaml  # Reference your environment.yml file
+        micromamba-version: "latest"
+        environment-file: environment.yaml
         init-shell: bash
         cache-environment: true
         post-cleanup: 'all'
-        # cache: true                        # Cache the micromamba environment
 
     - name: Install package
       shell: bash -l {0}
@@ -50,7 +49,6 @@ jobs:
     - name: Upload Documentation Artifact
       uses: actions/upload-pages-artifact@v3
       with:
-        # name: documentation
         path: doc/
 
   deploy:
@@ -64,14 +62,5 @@ jobs:
       name: github-pages
       url: ${{ steps.deployment.outputs.page_url }}
     steps:
-      # - name: Download Documentation Artifact
-      #   uses: actions/download-artifact@v4
-      #   with:
-      #     name: documentation
-      #     path: .
-
       - name: Deploy to GiHub Pages
         uses: actions/deploy-pages@v4
-        # with:
-        #   artifact_name: documentation
-

.github/workflows/run_tests.yaml

@@ -27,7 +27,7 @@ jobs:
       - name: Setup micromamba
         uses: mamba-org/setup-micromamba@v1
         with:
-          environment-file: environment_cpu.yaml
+          environment-file: ${{ runner.os == 'Windows' && 'environment_cpu_win.yaml' || 'environment.yaml' }}
           create-args: >-
             python=${{ matrix.python-version }}
 

.gitignore

@@ -12,3 +12,4 @@ scripts/cooper/training/copy_testset.py
 scripts/rizzoli/upsample_data.py
 scripts/cooper/training/find_rec_testset.py
 synapse-net-models/
+scripts/portal/upscale_tomo.py

README.md

@@ -2,7 +2,7 @@
 
 SynapseNet is a tool for segmentation and analysis of synapses in electron microscopy.
 
-To learn how to use SynapseNet, check out [the documentation](https://computational-cell-analytics.github.io/synapse-net/).
+To learn how to use SynapseNet, check out [the documentation](https://computational-cell-analytics.github.io/synapse-net/) and [the tutorial video](https://youtu.be/7n8Oq1uAByE).
 To learn more about how it works, check out [our preprint](https://www.biorxiv.org/content/10.1101/2024.12.02.626387v1).
 
 See an example reconstruction of a mossy fibre synapse with SynapseNet.

doc/.gitignore

@@ -0,0 +1 @@
+video/

doc/start_page.md

@@ -10,15 +10,15 @@ It provides deep neural networks for:
 
 It also offers functionality for quantifying synaptic ultrastructure based on segmentation results, for example by measuring vesicle or structure morphology, measuring distances between vesicles and structures, or assigning vesicles into different pools.
 SynapseNet mainly targets electron tomography, but can also be appled to other types of electron microscopy,
-especially throught the [domain adaptation](domain-adaptation) functionality.
+especially throught the [domain adaptation](#domain-adaptation) functionality.
 
-SynapseNet offers a [napari plugin](napari-plugin), [command line interface](command-line-interface), and [python library](python-library).
+SynapseNet offers a [napari plugin](#napari-plugin), [command line interface](#command-line-interface), and [python library](#python-library).
 Please cite our [bioRxiv preprint](https://www.biorxiv.org/content/10.1101/2024.12.02.626387v1) if you use it in your research.
 
 
 ## Requirements & Installation
 
-SynapseNet was developed and tested on Linux. It should be possible to install and use it on Mac or Windows, but we have not tested this.
+SynapseNet was developed and tested on Linux. It is possible to install and use it on Mac or Windows, but we have not extensively tested this.
 Furthermore, SynapseNet requires a GPU for segmentation of 3D volumes.
 
 You need a [conda](https://docs.conda.io/projects/conda/en/latest/user-guide/install/index.html) or [mamba](https://mamba.readthedocs.io/en/latest/installation/mamba-installation.html) installation. Follow the instruction at the respective links if you have installed neither. We assume you have `conda` for the rest of the instructions. After installing it, you can use the `conda` command.
@@ -36,6 +36,7 @@ cd synapse-net
 ```bash
 conda env create -f environment.yaml
 ```
+If you are using Windows then you have to use a different environment file for the installation: Use `environment_gpu_win.yaml` if you have a GPU and `environment_cpu_win.yaml` without a GPU.
 - You will need to confirm this step. It will take a while. Afterwards you can activate the environment:
 ```bash
 conda activate synapse-net
@@ -72,6 +73,8 @@ We currently offer seven different models for segmenting synaptic structures:
 
 ## Napari Plugin
 
+You can find a video tutorial for the SynapseNet napari plugin [on YouTube](https://youtu.be/7n8Oq1uAByE). Below, we explain the different plugin components with screenshots.
+
 After installing SynapseNet you can start napari by activating the `synapse-net` environment (or another environment you installed it in) and executing the `napari` command.
 Once napari is opened, you can load a tomogram (or other image data), by drag'n'dropping the corresponding mrc file onto the napari window.
 
@@ -110,7 +113,7 @@ The screenshot below shows a grouping of vesicles into 'close' (red) and 'far' (
 In addition, the `Segmentation Postprocessing` widget can be used to filter out objects that do not overlap with a mask, e.g. a synaptic compartment, or to intersect a segmentation with the boundaries of a mask.
 
 
-## Command Line Functionality
+## Command Line Interface
 
 SynapseNet provides a command line interface to segment synaptic structures in mrc files (or other image formats), and to export segmentation results to IMOD.
 
@@ -161,3 +164,10 @@ Domain adaptation is implemented in `synapse_net.training.domain_adaptation`. Yo
 
 We also provide functionality for 'regular' neural network training. In this case, you have to provide data **and** manual annotations for the structure(s) you want to segment.
 This functionality is implemented in `synapse_net.training.supervised_training`. You can find an example script that shows how to use it [here](https://github.com/computational-cell-analytics/synapse-net/blob/main/examples/network_training.py).
+
+## Segmentation for the CryoET Data Portal
+
+We have published segmentation results for tomograms of synapses stored in the [CryoET Data Portal](https://cryoetdataportal.czscience.com/). So far we have made the following depositions:
+- [CZCDP-10330](https://cryoetdataportal.czscience.com/depositions/10330): Contains synaptic vesicle segmentations for over 50 tomograms of synaptosomes. The segmentations were made with a model domain adapted to the synaptosome tomograms.
+
+The scripts for the submissions can be found in [scripts/cryo/cryo-et-portal](https://github.com/computational-cell-analytics/synapse-net/tree/main/scripts/cryo/cryo-et-portal).

environment.yaml

@@ -1,6 +1,4 @@
 channels:
-    - pytorch
-    - nvidia
     - conda-forge
 name:
     synapse-net
@@ -13,10 +11,10 @@ dependencies:
     - pyqt
     - python-elf
     - pytorch
-    - pytorch-cuda=12.4
     - tensorboard
     - torch_em
     - torchvision
     - trimesh
+    - zarr < 3
     - pip:
         - napari-skimage-regionprops

environment_cpu.yaml renamed to environment_cpu_win.yaml

@@ -4,15 +4,20 @@ channels:
 name:
     synapse-net
 dependencies:
+    - cpuonly
     - bioimageio.core
     - cpuonly
     - kornia
+    # This pin is necessary because later nifty versions have import errors on windows.
+    - nifty =1.2.1=*_4
     - magicgui
     - napari
+    - protobuf <5
     - pip
     - pyqt
     - python-elf
     - pytorch
+    - torchvision
     - tensorboard
     - torch_em
     - trimesh

environment_gpu_win.yaml

@@ -0,0 +1,25 @@
+channels:
+    - pytorch
+    - nvidia
+    - conda-forge
+name:
+    synapse-net
+dependencies:
+    - bioimageio.core
+    - kornia
+    # This pin is necessary because later nifty versions have import errors on windows.
+    - nifty =1.2.1=*_4
+    - magicgui
+    - napari
+    - protobuf <5
+    - pip
+    - pyqt
+    - python-elf
+    - pytorch
+    - pytorch-cuda>=11.7  # you may need to update the cuda version to match your system
+    - torchvision
+    - tensorboard
+    - torch_em
+    - trimesh
+    - pip:
+        - napari-skimage-regionprops

examples/analysis_pipeline.py

@@ -1,3 +1,6 @@
+import os
+
+import imageio.v3 as imageio
 import napari
 import pandas as pd
 import numpy as np
@@ -128,6 +131,13 @@ def save_analysis(segmentations, vesicle_attributes, save_path):
     vesicle_attributes.to_excel(save_path, index=False)
 
 
+def save_segmentations(segmentations):
+    output_folder = "segmentations"
+    os.makedirs(output_folder, exist_ok=True)
+    for name, segmentation in segmentations.items():
+        imageio.imwrite(os.path.join(output_folder, f"{name}.tif"), segmentation, compression="zlib")
+
+
 def main():
     """This script implements an example analysis pipeline with SynapseNet and applies it to a tomogram.
     Here, we analyze docked and non-attached vesicles in a sample tomogram."""
@@ -150,7 +160,11 @@ def main():
     vesicle_attributes = assign_vesicle_pools(vesicle_attributes)
 
     # Visualize the results.
-    visualize_results(tomogram, segmentations, vesicle_attributes)
+    # visualize_results(tomogram, segmentations, vesicle_attributes)
+
+    # Save the segmentation results to tif files so that they can be re-used later.
+    # They will be saved to the folder 'segmentations'.
+    save_segmentations(segmentations)
 
     # Compute the vesicle radii and combine and save all measurements.
     save_path = "analysis_results.xlsx"