Tidy genomic and transcriptomic single-cell analyses
Author(s): Stefano Mangiola,Michael I Love
Affiliation(s): WEHI
Social media: https://twitter.com/steman_research
This workshop will present how to perform genomic and transcriptomic data analysis using the tidy data paradigm. This paradigm became the standard in R data analysis across many fields. It provides a standard way to organise data values, where each variable is a column, each observation is a row, and data is manipulated using a familiar and easy-to-understand vocabulary. The data structure remains consistent across manipulation and analysis functions. Tidy genomic analyses, such as tests for genomic enrichment, can be performed with plyranges and nullranges. You will learn techniques for tidy manipulation of genomic range data; how to generate and compare to ranges representing a particular null hypothesis; how to decide between bootstrapping versus creating a matched set from the background; and how to perform more complex operations, such as computing correlations of activity at promoters and nearby enhancers. Tidy transcriptomic analyses can be performed with tidySingleCellExperiment, tidySummarisedExperiment, tidybulk and tidyverse. You will learn the differences between SingleCellExperiment, SummarizedExperiment and their tidy representation; basic tidy operations possible with tidySingleCellExperiment; how to manipulate and visualise SingleCellExperiment in a tidy manner with a real-world case study that will include single-cell and (pseudo-)bulk analyses.
Package demo details
Source code