I’m very pleased to announce the release of JBrowse 1.5.0! There are three headline features in this release. First, we’ve integrated a direct-access BigWig data backend, adapted from the Dalliance Genome Explorer that can read wiggle data directly from compressed BigWig files stored on your web server. Second, there is now a beautiful, full-featured canvas-based
Wiggle track type. Third, we have a very powerful new click system that allows JBrowse administrators complete flexibility in configuring what happens when a user left- or right-clicks a feature in an HTML-based feature track, including right-click context menus.
A demonstration of JBrowse 1.5.0 showing a test BigWig-based wiggle track (alongside an old-style image-based wiggle track for comparison) can be seen here, and a whole-genome RNA-seq profile of tomato, with corresponding gene models, can be seen here.
As usual, this release comes in two flavors: the “minimal” release (JBrowse-1.5.0-min.zip – 2.8M) that includes only the software and documentation necessary to format your own data and run the browser, and the “full” release (JBrowse-1.5.0-min.zip – 27.2M) that includes the developers’ test suite, more sample data files, and developer documentation.
Here is the full list of new improvements in 1.5.0:
Added a direct-access storage driver for BigWig data files, based
on code from the Dalliance Genome Explorer by Thomas Down. BigWig
file access is supported now by the current versions of all major
browsers except Internet Explorer (which is expected to work when
version 10 is released along with Windows 8).
canvas-based wiggle track implementation for quantitative
data that, when used with the new BigWig storage backend, removes
the need to pre-generate rendered images of wiggle data. Its
display is also highly configurable, with configuration options
modeled on the GBrowse
wiggle_xyplot glyph type
Added highly configurable behavior for left-clicking and
right-clicking features in HTML-based feature tracks. If a
menuTemplate option is specified in the track configuration,
right-clicking a feature brings up a context menu, the items in
which can be configured to do nearly anything, but that are easy to
configure for the very common use case of wanting to display
content from a certain URL. Feature left-clicks are also
configurable using the same mechanism. Thanks to Alexie
Papanicolaou and Temi Varghese for the initial implementation of
Improved the default HTML feature left-click dialog box. It is now
both prettier, and more comprehensive, displaying all available
data for the feature.
Added a small helper script,
add-track-json.pl that developers
and advanced users can use to programmatically add a block of track
configuration JSON to an existing JBrowse configuration file.
Improved / fixed vertical alignment of sub-elements of HTML
features, including subfeatures and the arrowheads that show
strand. All elements in a feature are now vertically centered by
JBrowse 1.4.2 has been released, containing some small fixes for issues in the 1.4.1 release.
Changes in this release:
- Restore support for histScale, subfeatureScale, and labelScale in
--clientConfig command-line options and track configuration JSON. Thanks to Hernán Bondino for pointing this out.
JBrowse 1.4.1 has been released, containing some small fixes for issues in the 1.4.0 release.
Changes in this release:
- Made displayColumns option for the faceted track selector
case-insensitive, and interpret a column name of “Name” as meaning
the track’s key.
bam-to-json.pl now filters out alignments that are not at least
two nucleotides in length. Thanks to Tristan Lubinski for
- Introduced limits on the sizes of cookies that can be set,
preventing ‘400 bad request’ errors (issue #113).
JBrowse 1.4.0 has been released! There are two headline features in this release: a new, extremely powerful faceted track selector that makes it easy for users to search through hundreds or thousands of tracks based on their metadata, and much better support for large numbers of reference sequences, enabled by retiring the old reference sequence dropdown selector in favor of adding a new find-as-you-type feature to the location box.
A demonstration of the faceted track selector can be seen here. This contains a snapshot of the modENCODE track metadata taken from http://data.modencode.org. Note that the track data and reference sequences in this track selection demo are not real: the reference sequences and wiggle data are all just copies of the volvox wiggle test track from the JBrowse test suite.
As is becoming the norm, this release comes in two flavors: the “minimal” release (JBrowse-1.4.0-min.zip – 2.6M) that includes only the software and documentation necessary to format your own data and run the browser, and the “full” release (JBrowse-1.4.0-full.zip – 5.5M) that includes the developers’ test suite, more sample data files, and developer documentation.
There have been lots of improvements over the 1.3 series:
- Added a full-featured faceted track selector for users that have
many (hundreds or thousands) of tracks. This can be turned on by
trackSelectorType config variable to “Faceted”. See
the JBrowse wiki for more documentation on how to use faceted track
selection. (issue #95)
- Removed the dropdown selector for reference sequences in favor of
making the location box auto-complete reference sequence and
feature names. This makes JBrowse much more scalable to large
numbers of reference sequences. (fixes issues #3, #60, and #101)
- Added a vertical-scrolling marker on the right side of the track
pane, making it much easier to discern the vertical position of the
track display. (issue #93).
- biodb-to-json.pl and flatfile-to-json.pl now load all available
feature data: all attributes of features are now encoded in the
JSON and are available for use by feature callbacks. (issue #72)
- Feature labels now do not scroll off screen if any part of the
feature is still visible (fixes issue #62).
- Added jbrowse_conf.json, a default JSON-format configuration file,
to the JBrowse root directory. Makes it easier to get started
with more advanced JBrowse configuration.
- JBrowse instances now report anonymous usage statistics to the JBrowse developers. This data is very important to the JBrowse project,
since it is used to make the case to grant agencies for continuing
to fund JBrowse development. No research data is transmitted, the
data collected is limited to standard Google Analytics, along with
a count of how many tracks the JBrowse instance has, how many
reference sequences are present, their average length, and what
types of tracks (wiggle, feature, etc) are present. Users can
disable usage statistics by setting “suppressUsageStatistics: true”
in the JBrowse configuration.
Oscar Westesson has contributed a nice demonstration of using JBrowse 1.2.1 to display and compare performance of four different XRate grammars on the poliovirus genome.
Download .zip (35M)
JBrowse 1.3.1 has been released, this is a minor release with fixes for some bugs found in 1.3.0.
- JBrowse-1.3.1-min.zip – 2.0MB
- Includes all that is needed for running JBrowse and formatting your own data.
- JBrowse-1.3.1-full.zip – 4.7MB
- Includes developer API documentation, developer test suites, additional sample data
Changes in this release
* Fixed memory-management bug that caused way too much RAM to be used
by FeatureTrack loading (flatfile, bam, and biodb-to-json.pl) when
loading with very large numbers of reference sequences. Big thanks
to Tristan Lubinski for help in isolating this.
* Fixed some bugs in BAM support section of setup.sh autosetup
script, thanks to Tristan Lubinski for help in isolating this as
* Added an example document with an iframe running JBrowse in
embedded mode in docs/examples/embedded_mode.html
* flatfile-to-json.pl now loads the ‘score’ attribute of features in the
JBrowse version 1.3.0 has arrived!
We have a long list of improvements in this release, some of which have been a long time coming. It’s full speed ahead from here: more major new features are just around the corner!
This release comes in two flavors: the “minimal” release (JBrowse-1.3.0-min.zip – 2.0M) that includes only the software and documentation necessary to format your own data and run the browser, and the “full” release (JBrowse-1.3.0-full.zip – 4.7M) that includes the developers’ test suite, more sample data files, and developer documentation.
One important note about browser support: as of this 1.3.0 release, JBrowse no longer supports Microsoft Internet Explorer 6.
And now, the Big List of New Features!
This is a minor bugfix release, containing fixes for display of unstranded features and a correction to the tutorial HTML file that incorrectly suggested using flatfile-to-json.pl to format feature data from a BAM file.
JBrowse 1.2.2 can be downloaded from http://jbrowse.org/releases/jbrowse-1.2.2.zip
A new tar.gz file is available, containing a copy of JBrowse 1.2.1 with Human hg19 data. Being based on a more recent version of JBrowse, it should work much better than the previous hg19 tarball releases. Enjoy!
The new file is here: http://jbrowse.org/releases/jbrowse-1.2.1-hg19mini.tar.gz
Great news! JBrowse has been accepted to Google Summer of Code 2012 under the Genome Informatics umbrella organization. Many thanks to Reactome’s Robin Haw for shepherding the Genome Informatics organization through the application process!
Are you a college undergraduate interested in working on JBrowse this summer? You should consider applying to Google Summer of Code to fund you for the summer! The student application deadline is April 6, 2012.