JBrowse getting started guide

Getting Started With JBrowse

This page should be enough to get you started. If you run into trouble, please join the mailing list and ask us about it.

Prerequisites

From CPAN:

BioPerl (version 1.6)
JSON (version 2)
JSON::XS (optional, for speed)

Download/install

Get JBrowse from GitHub (click on the "download" button)

Then unzip/untar JBrowse in a directory that's served by apache (e.g., /var/www/html/)

Reference Sequences

Tell JBrowse about your reference sequences with bin/prepare-refseqs.pl

(it's convenient to run the JBrowse scripts from within the JBrowse root directory)

$ bin/prepare-refseqs.pl --fasta docs/tutorial/data_files/volvox.fa

Features

There are two options for telling JBrowse about your feature data: if you have a genomic annotation database such as chado, Bio::DB::SeqFeature::Store, or Bio::DB::GFF, then you can use JBrowse's biodb-to-json.pl. On the other hand, if you mainly deal with flat files like GFF or BED, it may be more convenient to use JBrowse's flatfile-to-json.pl.

biodb-to-json.pl -

You use bin/biodb-to-json.pl with a configuration file (documented here).

$ bin/biodb-to-json.pl --conf docs/tutorial/conf_files/volvox.json

bin/flatfile-to-json.pl -

bin/flatfile-to-json.pl doesn't use a configuration file; instead, per-track settings are given as command-line arguments. Run bin/flatfile-to-json.pl without any arguments to see a description of the available settings. Most of them correspond to track settings in the config file, so see the track settings section in the config file documentation for more details.

Names

To enable users to search using feature names or IDs, you have to do two things:

specify autocomplete for the tracks whose features you want to be searchable (see the track settings section in the config file documentation for details about the autocomplete setting). When you do this, a file with a list of names and locations will be generated for each track that has the autocomplete setting.
run bin/generate-names.pl to collect those per-track files together into the structure that JBrowse uses to search

$ bin/generate-names.pl -v

You need to run bin/generate-names.pl again every time you add new features to JBrowse.

And that's all, for basic usage.

visit the main page to view the results.

Quantitative tracks

If you want to browse quantitative ("wiggle") tracks in JBrowse, you need to do a bit more work.

Prerequisites:

libpng

Compile the wiggle-processing program by running make in the JBrowse root directory

$ make
g++   -O3 -lpng -o bin/wig2png src/wig2png.cc
$

System-specific:

Apple OS X:

You need libpng and a C++ compiler; you can get these from MacPorts or Fink.

Once you have those installed, you need to compile the JBrowse wiggle-processing program. JBrowse includes a makefile to do this, but you may need to add the paths for libpng and png.h to your compiler's library and include paths. For example, if libpng is in /opt/local, you can run make like this:
```
$ make GCC_LIB_ARGS=-L/opt/local/lib GCC_INC_ARGS=-I/opt/local/include
```
libpng might be somewhere other than /opt/local, like /usr/X11R6 for example. To find where libpng is, try locate libpng

Red Hat/CentOS/Fedora:
install the libpng and libpng-devel packages. (The rest should just work)

Ubuntu:
install the libpng12 and libpng12-dev packages. (The rest should just work)

Once wig2png is compiled, you run it via a wrapper script:

$ bin/wig-to-json.pl --wig docs/tutorial/data_files/volvox_microarray.wig