GENBAIT is a Python package designed for bait (feature) selection in proximity labeling data using genetic algorithms.
A preprint describing the method and introducing a novel benchmarking platform is available: Kasmaeifar et al. (2024) Computational design and evaluation of optimal bait sets for scalable proximity proteomics
GENBAIT requires Python 3.10 or higher. We recommend creating a virtual environment to ensure smooth installation.
To install GENBAIT, you first need Git. Follow the instructions below to install Git on your system.
-
Download the Git installer:
- Go to the official Git website: https://git-scm.com/download/win.
- Download the latest installer for Windows.
-
Run the installer:
- Locate the downloaded file and double-click to open the installer.
- Follow the prompts in the setup wizard. You can keep the default options or customize the installation.
-
Install Git using Homebrew:
- If you have Homebrew installed, open the Terminal and run:
brew install git
- Homebrew will handle the download and installation of Git.
- If you have Homebrew installed, open the Terminal and run:
-
Verify the installation:
- In Terminal, type the following command and press Enter:
git --version
- You should see a Git version number, confirming that Git is installed.
- In Terminal, type the following command and press Enter:
To install the genbait
package, follow these steps:
-
Install GENBAIT from GitHub using pip:
pip install git+https://github.com/camlab-bioml/genbait.git
-
Ensure all dependencies are installed: GENBAIT requires the following Python packages:
- pandas
- numpy
- scipy
- scikit-learn
- matplotlib
- seaborn
- gprofiler-official
- igraph
- leidenalg
- deap (for Genetic Algorithm operations)
These packages will be installed automatically during the setup.
Installation takes less than 2 minutes.
A detailed tutorial of how to use different functions of the package can be found here: GENBAIT Tutorial
For 200 baits and 10 iterations for a panel size 50, running genbait takes approximately 30 minutes on a computer with 32 GB RAM.
This software is authored by: Vesal Kasmaeifar, Kieran R Campbell
Lunenfeld-Tanenbaum Research Institute & University of Toronto