If you are new to Scaden, plesase consider reading the paper prior to use.
If you rather use Scaden locally, see the package quick start section below, or read the package documentation.
Scaden (Single-cell Assisted Deconvolutional Network) is a tool for bulk RNA-seq cell type deconvolution that uses an ensemble of deep neural networks trained on artificial bulk data. The bulk data is simulated from scRNA-seq datasets (see figure below). This method was developed in the Genome Biology of Neurodegenerativ Diseases group at the DZNE Tübingen and the Medical Systems Biology group at the ZMNH .
The pre-print describing Scaden is available on Biorxiv.
Overview of training data generation and cell type deconvolution with Scaden. A: 135 Artificial bulk samples are generated by subsampling random cells from a scRNA - seq datasets 136 and merging their expression profiles. B: Model training and parameter optimization on 137 simulated tissue RNA - seq data by comparing cell fraction predictions to ground - truth cell 138 composition. C: Cell deconvolution of real tissue RNA - seq data using Scaden.
Scaden be easily installed on a Linux system, and should also work on Mac. There are currently two options for installing Scaden, either using Bioconda or via pip.
If you don't want to install Scaden at all, but rather use a Docker container, we provide that as well. You can pull the Scaden docker container with the following command (from Dockerhub):
For a typical deconvolution with Scaden you will have to perform three steps:
- pre-processing of training data
- training of Scaden model
This assumes that you already have a training dataset. If not, Scaden contains functionality to create a dataset from one or several scRNA-seq datasets
Example data can be found at https://scaden.ims.bio/downloads/