iSEE: Interactive visualization of SummarizedExperiment objects

Instructor(s) name(s) and contact information

• Charlotte Soneson ([email protected])
• Kevin Rue-Albrecht ([email protected])
• Federico Marini ([email protected])
• Aaron Lun ([email protected])

Key resources

• Docker image: Docker Hub
• Workshop: pgkdown website

Workshop Description

This workshop demonstrates the use of the iSEE package to create and configure interactive applications for the exploration of various types of genomics data sets (e.g., bulk and single-cell RNA-seq, CyTOF, gene expression microarray).

This workshop will be presented as a lab session that combines an instructor-led live demo, followed by hands-on experimentation guided by completely worked examples and stand-alone notes that participants may continue to use after the workshop.

The instructor-led live demo comprises three parts:

1. Brief lecture on the package concept and functionality
2. Overview of the graphical user interface
3. Instructions to preconfigure iSEE apps

The hands-on lab comprises three parts:

1. Inspection of single-cell RNA-seq data at various steps of a typical computational workflow, including quality control and dimensionality reduction
2. Addition of custom panels to the user interface for advanced visualization.
3. Additional questions from the participants, including individual use cases and suggestions for future developments

Participants are encouraged to ask questions at any time during the workshop.

Pre-requisites

• Basic knowledge of R syntax and the use of data-frames
• Familiarity with the SummarizedExperiment and SingleCellExperiment classes
• Familiarity with the shiny CRAN package
• Familiarity with the TENxPBMCData package and vignette

Additional background reading about the programming environment, relevant packages, and example use cases:

• Shiny from RStudio: https://shiny.rstudio.com
• SummarizedExperiment paper: https://www.nature.com/articles/nmeth.3252 (Figure 2)
• iSEE paper: https://f1000research.com/articles/7-741/v1
• "Orchestrating single-cell analysis with Bioconductor", https://www.nature.com/articles/s41592-019-0654-x, Nat Methods 17, 137–145 (2020)
  • https://osca.bioconductor.org, as an online companion to the manuscript above
• The iSEEu package for extending the functionality of iSEE

Workshop Participation

Students will participate by following along an R markdown document, and asking questions throughout the workshop. There is also scope for participants to apply iSEE to their own data sets, and fuel the discussion with more questions about specific use cases.

R / Bioconductor packages used

1. iSEE : https://bioconductor.org/packages/iSEE
2. SummarizedExperiment: https://bioconductor.org/packages/SummarizedExperiment
3. SingleCellExperiment: https://bioconductor.org/packages/SingleCellExperiment
4. scater: https://bioconductor.org/packages/scater
5. scran: https://bioconductor.org/packages/scran
6. TENxPBMCData: https://bioconductor.org/packages/TENxPBMCData
7. BiocSingular: https://bioconductor.org/packages/BiocSingular

Time outline

Activity	Time
Lecture: Overview of package and concepts	15m
Live demo: the user interface	15m
Lab: Configuring the app interface	20m
Lab: A single-cell RNA-seq workflow	20m
Lab: Custom panels and advanced features	20m
Additional questions	15m

Total: 1h45

Workshop goals and objectives

Learning goals

• Recognize the benefits of integrative data containers such as SummarizedExperiment and SingleCellExperiment for downstream analyses and visualization
• Outline the unique features of iSEE built upon the RStudio Shiny framework
• Identify biological data that may be combined into insightful graphical outputs
• Utilize interactive GUI components and layouts to efficiently extract information from biological data sets
• Describe how to construct interactive apps and custom panels

Learning objectives

• Memorize the key information available in SummarizedExperiment and SingleCellExperiment objects
• Set up a local environment for running iSEE apps
• Interact with components of the iSEE user interface to visually inspect and discuss various data sets
• Identify and locate configurable aspects of iSEE apps
• Practice interactive visualization over a single-cell RNA-sequencing workflow
• Design custom iSEE panels for advanced use cases
• Imagine use cases and future developments for interactive visualization as part of computational workflows

Docker setup

In the Docker settings menu:

• Open 'Preferences...'
• Navigate to 'Resources'
• Use the slider to set 'Memory' to '4.00 GB'
• Click 'Apply & Restart'

Workshop setup

• Run docker pull iseedevelopers/iseeworkshop2020
• Run docker run -e PASSWORD=isee -p 8787:8787 -d --rm iseedevelopers/iseeworkshop2020. Use -v $(pwd):/home/rstudio argument to map your local directory to the container.
• Log in to RStudio at http://localhost:8787 using username rstudio and password isee. Note that on Windows you need to provide your localhost IP address like http://191.163.92.108:8787/ - find it using docker-machine ip default in Docker's terminal.
• Run browseVignettes(package = "iSEEWorkshop2020"). Click on one of the links, "HTML", "source", "R code".
  • In case of The requested page was not found error, add help/ to the URL right after the hostname, e.g., http://localhost:8787/help/library/iSEEWorkshop2020/doc/iSEE-lab.html. This is a known bug.