Advanced JSON tutorial

JBrowseR() takes JBrowse’s createLinearGenomeView options as named arguments, and JBrowseRApp() takes createApp’s, both under the camelCase names JBrowse gives them. A browser can therefore live in a JSON file that R, Python and JavaScript all read.

library(JBrowseR)

What the options hold

JBrowseR()’s options are assembly, tracks, location, session, aggregateTextSearchAdapters, internetAccounts, plugins and configuration. The package passes them through unread, so an option JBrowse adds works here without a package update.

A JBrowse Web config.json is not that shape: it lists assemblies in the plural and opens a defaultSession. Its assemblies, tracks and plugins are JBrowseRApp() options, though, and its defaultSession.views is JBrowseRApp()’s views. To open a genome JBrowse hosts, name its hub instead — JBrowseR(assembly = "hg19") fetches https://jbrowse.org/ucsc/hg19/config.json and resolves its paths.

From a JSON file

do.call() hands a list’s entries to a function as named arguments, so a file read with jsonlite::read_json() is a whole browser. The load_config_json example app keeps its assembly in one:

do.call(JBrowseR, jsonlite::read_json("config.json"))

modifyList() overrides a field the file holds:

opts <- jsonlite::read_json("config.json")
do.call(JBrowseR, modifyList(opts, list(location = "10:29,838,737..29,838,819")))

read_json() keeps JSON arrays as lists, so a one-element assemblyNames stays an array on the way back out.

From an R list

The same object written inline:

opts <- list(
  assembly = list(
    name = "hg19",
    uri = "https://jbrowse.org/genomes/hg19/fasta/hg19.fa.gz",
    aliases = list("GRCh37")
  ),
  tracks = list(
    list(
      uri = "https://jbrowse.org/genomes/hg19/GRCh37_latest_genomic.sort.gff.gz",
      name = "NCBI RefSeq Genes"
    )
  ),
  location = "10:29,838,737..29,838,819"
)

do.call(JBrowseR, opts)