Chapter 1 Introduction

This introductory course will provide a quick overview of how the Bayesian NMF algorithm, CoGAPS (Coordinated Gene Activity across Pattern Subsets), can provide new insights into single cell datasets. Through these exercises you will analyze a real dataset using the SciServer compute platform.

1.1 Motivation

If you would like to perform sparse matrix factorization on any data. And when this data represents biomolecules, to do gene set analysis. This can be done with CoGAPS, which can be used by anyone; no machine learning experience is required.

1.2 Target Audience

The course is intended for anyone! No software or prior coding experience is required.

1.3 Curriculum

The course covers:

  • How to join the compute platform, SciServer
  • How to access and launch cellxgene
  • How to load packages, data, configure/run CoGAPS, visualize patterns, find pattern markers, and document software in RStudio