New Tutorial Suite: Deconvolution with MuSiC, from public data to disease interrogation!


Posted on: 29 November 2022

The still new and shiny single-cell analysis topic now boasts a deconvolution tutorial suite! What does deconvolution do you ask? Well, in this context, it infers cell proportions from bulk RNA-seq data. You heard that correctly - instead of expensive new single-cell experiments, you can re-analyse old bulk RNA-seq data and estimate cell proportions. All you need is a reasonably good single cell dataset to use as a reference and you’re good to go! The tutorial suite shows you how to build your reference from publicly available single cell data, and apply analysis to some publicly available bulk RNA-seq data.

This suite was built byorcid logoAvatarWendi Bacon,AvatarMehmet Tekmanand we’d love to hear your thoughts, so do fill out the feedback form at the ends or give us a shout on the galaxy-single-cell Gitter/Matrix forums!

We are also planning to test this as potential undergraduate capstone projects, so please get in touch if you’d like to be a help with such a study.

Heatmap with columns of cell types vs rows of inferred and actual cell type proportions for A and B samples. Cell colours are similar between actual and inferred.