Unsupervised data integration

This is a bonus exercise, for those of you who are interested and have some time over. It is from the course Omics Integration and Systems Biology, and the lab was created by Nikolay Oskolkov. In this exervise you will analyze a single cell data set with chromatin accessebility (scATACseq), DNA methylation (scBSseq) and gene expression (scRNAseq) data, using MOFA.

Note that this tutorial only serves as a guide and not meant to be taken as a copy-and-paste exercise. The tutorial will give you a general framework in how to use MOFA as a tool to integrate multidimensional data and how to interpret MOFA’s results.

Learning outcomes

• apply feature selection methods on multi -omics data

• apply an unsupervised method for data integration, and interpret the results

Setup

This has been tested on Uppmax. First load these modules

module load bioinfo-tools
module load R/4.0.0
module load R_packages/4.0.0

You also need to install some python code that MOFA needs.

pip install mofapy

Then, get the data for the exercise, and go to the data directory

git clone https://github.com/NikolayOskolkov/UnsupervisedOMICsIntegration.git
cd UnsupervisedOMICsIntegration

Now you are ready to start R or rstudio.

The exercise

The instructions for the exercise are here. First, there is short introduction to MOFA. For the exercise, scroll down to “Prepare scNMT Data Set for MOFA”.