Skip to contents

HCABiocTraining is an introduction to Human Cell Atlas data retrieval and analysis in R / Bioconductor. The focus is on single-cell RNA-seq data. A wide range of material is covered, from basic R to advanced data analysis steps outlined in ‘Orchestrating Single-Cell Analysis with Bioconductor’ (OSCA). The treatment is very superficial, but the hope is that it unlocks opportunities for reproducible, sophisticated analysis.

Installation

This resource includes

Install necessary software from GitHub with:

if (!"BiocManager" %in% rownames(installed.packages()))
    install.packages("BiocManager", repos = "https://cran.r-project.org")
BiocManager::install("mtmorgan/HCABiocTraining", dependencies = TRUE)

The software makes use of the anndata Python module through the R package reticulate. Install anndata following reticulate best practices, e.g., by creating and using a virtual environment.

library(reticulate)
reticulate::virtualenv_create('HCABiocTraining', packages = 'anndata')
reticulate::use_virtualenv('HCABiocTraining')

Articles

The articles in this repository cover the following topics

  1. Introduction to R. Scripting, core functions, and contributed packages.

  2. HCA Data Access. Using the hca and cellxgenedp packages for data retrieval.

  3. Single Cell Sequence Analysis. Explores the Seurat and SingleCellExperiment frameworks for well-established single cell analysis work flows such as the Seurat documentation and R / Bioconductor ‘Orchestrating Single-Cell Analysis with Bioconductor’ (OSCA).

  4. Integrating single cell data into Seurat and Bioconductor workflows.

  5. Prospects for multimodal, spatial and extended analysis.

Most of our time will be spent on HCA data access and integration of single cell data into Seurat and Bioconductor workflows.

Docker Image

To make use of the docker image (after the workshop) pull it

docker pull ghcr.io/mtmorgan/hcabioctraining:latest

and run it using -p to connect the local port ‘8787’ to the port where RStudio is, and -v to mount your local system folder working_directory to the docker image

docker run -it --rm \
    -p 8787:8787 \
    -v /your/file/path/working_directory:/home/rstudio/working_directory

RStudio should be available in your browser at https://localhost:8787 or https://127.0.0.1:8787