viewmastR is a R framework for genomic cell type classification using the Burn machine learning library and its modules. viewmastR is a very flexible and customizable platform for labelling cell types in your data
The main features of viewmastR are:
- Use a blazingly fast machine learning approach to cell classification according to a reference dataset
- Augment data for rare cell types
- Classify single cell profiles according to a reference of bulk data
Additional features include:
- Classification of single-cell data using bulk reference datasets
- Super efficient bulk RNAseq deconvolution with intercept value
Installation
First you need to have an updated Rust installation. Go to this site to learn how to install Rust.
To install development version of viewmastR:
remotes::install_github("furlan-lab/viewmastR")How to start
We have a few vignettes for
-
vignette("HowTo")to explore the basics of the package. -
vignette("Augment")to explore data augmentation -
vignette("BulkClassify")to see how to use a bulk dataset to classify single-cell profiles -
vignette("Deconvolute")to see how to deconvolute a bulk dataset using single-cell profiles
