Skip to main content

JBrowseR

JBrowseR renders a JBrowse 2 linear genome view, drawn on the GPU, as an htmlwidget. Embed a full genome browser in an R Markdown document or Shiny app, or launch one from the R console. It shares the same framework-agnostic view core as the Python anywidget, so both stay in step.

Install

# install.packages("devtools")
devtools::install_github("GMOD/JBrowseR")

A declarative API

Every argument is a JBrowse embedding option, named as JBrowse names it and passed through unchanged. Assemblies, tracks and views are JBrowse's own config objects written as R lists, so any track type, view type or option works with nothing added to the package, and a whole config is do.call(JBrowseRApp, config).

Name a hosted genome and the assembly, reference-name aliases, cytobands, and gene-name search all come preconfigured, and location can be a gene symbol:

library(JBrowseR)

JBrowseR(assembly = "hg38", location = "BRCA1")

Add tracks by URL. The track type and index files (.bai/.crai/.tbi) are inferred from the extension:

JBrowseR(
  assembly = "hg38",
  tracks = list(
    list(
      uri = "https://jbrowse.org/genomes/GRCh38/alignments/NA12878/NA12878.alt_bwamem_GRCh38DH.20150826.CEU.exome.cram",
      name = "NA12878 Exome"
    )
  ),
  location = "17:43,044,295..43,048,000"
)

track_data_frame() turns an in-memory data frame into a track with no file or server, which is the one thing a config file can't express. A genome of your own is list(name = "mygenome", uri = "https://.../mygenome.fa.gz").

Comparing genomes

JBrowseR() shows a single linear genome view. JBrowseRApp() drives the full app from a views list, each entry a view as a config's defaultSession.views writes it, so a linear synteny view or a dotplot is one call:

base <- "https://jbrowse.org/demos/ecoli_pangenome"
strains <- c("K12", "Sakai", "CFT073", "NCTC86")

JBrowseRApp(
  assemblies = list(
    list(name = "K12", uri = paste0(base, "/K12.fa.gz")),
    list(name = "Sakai", uri = paste0(base, "/Sakai.fa.gz"))
  ),
  tracks = list(
    list(
      type = "SyntenyTrack",
      trackId = "ecoli_ava",
      name = "E. coli all-vs-all",
      assemblyNames = as.list(strains),
      adapter = list(
        type = "AllVsAllPAFAdapter",
        assemblyNames = as.list(strains),
        pafLocation = list(uri = paste0(base, "/all_vs_all.paf.gz"))
      )
    )
  ),
  views = list(
    list(
      type = "DotplotView",
      views = list(list(assembly = "K12"), list(assembly = "Sakai")),
      tracks = list("ecoli_ava")
    )
  )
)

The comparative-synteny vignette stacks four strains from the same alignment in a linear synteny view, the hosted data of the all-vs-all synteny tutorial.

Reacting to clicks in Shiny

Rendered inside Shiny, clicking a feature sets input$<outputId>_selected_feature to the feature's data, so tables, plots, and links can follow the current selection. Pair JBrowseROutput() with renderJBrowseR(), or JBrowseRAppOutput() with renderJBrowseRApp(), and update_jbrowse() changes options on a browser already on the page.

Run in Colab

A runnable R-runtime Colab notebook walks through the one-line genome, alignments, an R data-frame track, and cancer structural variants.

Full documentation is at gmod.github.io/JBrowseR.

See also