Installation#

Follow these steps to install VASCilia.

Requirements#

torch==1.12.1+cu113
torchvision==0.13.1+cu113
torchaudio==0.12.1+cu113
segmentation-models-pytorch
opencv-python
matplotlib
imagecodecs
tifffile
napari[all]
scikit-learn
readlif
czitools==0.4.1
czifile
npe2
numpy==1.26.4
colormap==1.1.0

Steps#

Set Up WSL#

Follow these steps to set up Windows Subsystem for Linux (WSL) with the Ubuntu 20.04 distribution.

Install WSL: 1. Open the Command Prompt and install the Ubuntu 20.04 distribution by copying and pasting the following command:
wsl --install -d Ubuntu-20.04
After the setup successfully completes, reboot your computer.
1. Open Ubuntu by typing “Ubuntu” in the Windows search bar.
Verify Installation: To check if CUDA and the GPU are correctly installed and available, type the following command in the Ubuntu terminal:
```
nvidia-smi
```
This should display information about your GPU and CUDA installation.

STEP 2: Download the Deep Learning Trained Models#

Download the VASCilia_trained_models from the following link: VASCilia Trained Models on Dropbox

After downloading, you should have a folder structure like this:

📁 models [Trained models]
├── 📁 cell_type_identification_model
│      Contains weights for cell type identification (IHC vs OHC).
├── 📁 new_seg_model
│      Where fine-tuned models will be stored.
├── 📁 region_prediction
│      Contains weights for region prediction.
├── 📁 seg_model
│      Contains weights for the 3D instance segmentation model.
├── 📁 Train_predict_stereocilia_exe
│      Executable for segmentation and retraining the model using WSL.
├── 📁 ZFT_trim_model
│      Contains deep learning model weights for the z-focus tracker algorithm.
└── 📁 rotation_correction_model
       Contains deep learning model weights for correcting stack orientation.

STEP 3: Download a Test Dataset#

Download one of our sample datasets to test VASCilia: Sample Datasets on Dropbox

After downloading, create a folder named raw_data and place the dataset inside it. Your folder structure should look like this:

📁 raw_data [Raw data (stacks) is placed here]
└── 📄 Litter 12 Mouse 4 MIDDLE - delBUNdelCAP_Airyscan Processing.czi

Create another folder named processed_data. This is where the plugin will store the analysis results.
```
📁 processed_data [Processed data will be stored here]
```

Instructions for Cloning and Installing the Repository#

You can set up VASCilia by following Option A or Option B:

Option A: Cloning the Repository#

Clone the repository:

git clone https://github.com/ucsdmanorlab/Napari-VASCilia.git
cd Napari-VASCilia

Create and activate the conda environment:

conda create -y -n napari-VASCilia -c conda-forge python=3.10
conda activate napari-VASCilia

Install dependencies:

pip install -r requirements.txt
pip install -e .

Launch Napari:
```
napari
```

Option B: Installing via PyPI#

Create and activate the conda environment:

conda create -y -n napari-VASCilia -c conda-forge python=3.10
conda activate napari-VASCilia

Download the requirements.txt file from this repository and ensure it is in your working directory.

Install dependencies and VASCilia:

pip install -r requirements.txt
pip install Napari-VASCilia

Launch Napari:
```
napari
```

Update the paths in config.json as needed. The config.json file will be generated upon running the plugin for the first time. The folder structure will look like this:

📁 C:/Users/Username/ [Your home folder]
├── 📁 .napari-vascilia [Folder path]
└── 📄 config.json

2. Update the `config.json` File: Edit the config.json file to reflect your system’s paths. Replace /…/ portions with the correct paths for your system. Example:

{
    "rootfolder": "C:/Users/.../processed_data/",
    "wsl_executable": "C:/Users/.../models/Train_predict_stereocilia_exe/Train_Predict_stereocilia_exe_v2",
    "model": "C:/Users/.../models/seg_model/stereocilia_v7/",
    "model_output_path": "C:/Users/.../models/new_seg_model/stereocilia_v8/",
    "model_region_prediction": "C:/Users/.../models/region_prediction/resnet50_best_checkpoint_resnet50_balancedclass.pth",
    "model_celltype_identification": "C:/Users/.../models/cell_type_identification_model/",
    "ZFT_trim_model": "C:/Users/.../models/ZFT_trim_model/",
    "rotation_correction_model": "C:/Users/.../models/rotation_correction_model/",
    "green_channel": 0,
    "red_channel": 1,
    "blue_channel": -1,
    "signal_intensity_channel": 0,
    "flag_to_resize": false,
    "flag_to_pad": false,
    "resize_dimension": 1200,
    "pad_dimension": 1500,
    "button_width": 100,
    "button_height": 35
}

3. Congratulations! 🎉 You are now ready to use VASCilia. Enjoy working with the plugin!

Multi-Batch Processing Feature: Required File#

The Multi-Batch Processing feature in this package requires an additional file: track_me_SORT_v3_exe.exe. This file is not included in the repository or the pip installation due to size constraints.

Download the File#

You can download the file from the following link:

Download track_me_SORT_v3_exe.exe

Instructions#

# If You Clone the Repository 1. Download the file from the link above. 2. Place the file in the following directory within the cloned repository:

src/napari_vascilia/core/

# If You Installed the Package via pip 1. Download the file from the link above. 2. Locate the installation directory for the package. To find the installation path, run the following Python code:

import napari_vascilia
print(napari_vascilia.__file__)

Place the downloaded file in the following directory:

Note: All other features of the package will function as expected without this file. This file is exclusively required for batch processing of multiple files.

Installation

Contents