Marker identification analysis workshop

Overview

This workshop will focus on performing marker identification analysis of transcriptomic data and visualising the results using mastR and smartid packages. We will identify markers on bulk RNA-seq data with mastR and single-cell RNA-seq data with smartid to explore feature importance in class and individual samples. Following this, we will perform gene-set scoring using tools from the smartid package. Overall, we will demonstrate a streamlined marker identification workflow for bulk data using mastR and a TF-IDF-based approach to process the scRNA-seq data, identify the markers, visualise and interpret resutls using smartid package.

The workshop will be organised into two broad sections:

• Identify markers on bulk RNA-seq data using mastR
  • Build markers pool from multiple sources
  • Perform differential expression analysis
  • Select signature from the pool
  • Refine signature by background expression
  • Visualize signature performance
• Identify markers on single-cell RNA-seq data using smartid
  • Calculate score for each feature in each sample
  • Scale and transform scores in regard of class
  • Identify markers for each class based on GMM
  • Perform gene-set scoring for validation

Detailed material can be found here and here.

Pre-requisites

The course is aimed at PhD students, Master's students, and third & fourth year undergraduate students. Some basic R knowledge is assumed - this is not an introduction to R course. If you are not familiar with the R statistical programming language it is compulsory that you work through an introductory R course before you attend this workshop.

R packages used

The following key R packages will be used:

• mastR
• smartid
• mclust
• edgeR
• limma

Time outline

Activity	Time
Introduction & setup	10m
Part 1. Identify markers on bulk RNA-seq data using mastR	35m
Part 2. Identify markers on single-cell RNA-seq data using smartid	35m
Q & A	10m

Workshop goals and objectives

Learning goals

• Understand the importance of marker identification in transcriptomic data analysis.
• Learn how to perform marker identification on bulk and scRNA-seq data in R.
• Understand the challenges caused by background/ambient expression within bulk data and rare population within scRNA-seq data.

Learning objectives

• Perform a marker identification analysis and interpret the results.
• Apply mastR to remove the confounding effects of background expression from the identification markers.
• Apply smartid to identify highly-specific markers for rare populations and to validate the results using scoring method in smartid.

Workshop package installation

Guide

This is necessary in order to reproduce the code shown in the workshop. The workshop is designed for R 4.5 and can be installed using one of the two ways below.

Via Docker image

If you're familiar with Docker you could use the Docker image which has all the software pre-configured to the correct versions.

docker run -e PASSWORD=password -p 8787:8787 gene233/markeridentificationworkflow:latest

Once running, navigate to http://localhost:8787/ and then login with Username:rstudio and Password:password.

You should see the Rmarkdown file with all the workshop code which you can run.

Via GitHub

Alternatively, you could install the workshop using the commands below in R 4.5.

install.packages('remotes')

# Install workshop package
remotes::install_github("Gene233/MarkerIdentificationWorkflow", build_vignettes = TRUE)

# To view vignettes
library(MarkerIdentificationWorkflow)
browseVignettes("MarkerIdentificationWorkflow")