scProjection
Projection and Deconvolution using deep hierarchical and generative neural network. Refer to our preprint: https://www.biorxiv.org/content/10.1101/2022.04.26.489628v1
Tutorials
First follow the install instructions below, at the bottom of the page, before following the tutorials.
Tutorial: Deconvolution of CellBench mixtures
Install scProjection
pip3 install scProjection
Package requirements
scProjection requires: Python 3. This is a guide to installing python on different operating systems.
(Python)
All platforms:
Alternative (On Windows):
- Download Python 3
- Make sure pip is included in the installation.
Alternative (On Ubuntu):
- sudo apt update
- sudo apt install python3-dev python3-pip
Alternative (On MacOS):
- /usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
- export PATH="/usr/local/bin:/usr/local/sbin:$PATH"
- brew update
- brew install python # Python 3
Setup of virtualenv
scProjection also requires: tensorflow, tensorflow-probability, sklearn and numpy. It is generally easier to setup the dependencies using a virtual environment which can be done as follows:
## Create the virtual environment
virtualenv -p python3 pyvTf2
## Launch the virtual environment
source ./pyvTf2/bin/activate
## Setup dependencies
pip3 install tensorflow
pip3 install tensorflow-probability
pip3 install sklearn
pip3 install numpy
## Install scProjection
pip3 install scProjection