Announcements

The Genomic HyperBrowser Jan 2016 update

posted Jan 30, 2016, 8:26 AM by Sveinung Gundersen

Dear users of the Genomic HyperBrowser,

Here is a small newsletter to inform you of current and future updates from the Genomic HyperBrowser project. The Genomic HyperBrowser is a framework for finding and testing relations between high-throughput sequencing and/or other genomic datasets, making use of statistical hypothesis testing and other advanced methodology. 


GSuite Tools

------------

The HyperBrowser project, which started 2007 as a collaboration between UiO and the Norwegian Radium Hospital (OUS), has in the recent year undergone dramatic improvements focused on epigenome-wide analyses, making powerful use of user-specified suites, or collections, of related datasets. This new expansion, called GSuite Tools, empowers researcher to ask questions about their genomic datasets in relation to the vast amount of datasets for different cell types/tissues or different epigenomic marks that has been made available from international projects like ENCODE or Roadmap Epigenomics. GSuite Tools contains user-friendly guides to answer common domain-specific questions and includes tools for defining suites of datasets, bulk downloading and analysis. GSuite Tools is not yet public, but we still welcome research collaborations, making use of the in-house versions of the software.


Moving to GitHub

----------------

We have just created a source code repository in GitHub for the Genomic HyperBrowser project. The aim is to make it easier to install the HyperBrowser, as well as to make it easier to include updates from the core Galaxy framework. In addition, we aim to better follow standard Open Source practices by having a public source code for transparent development, issue tracking, and support for developers and users. To provide this, we have now created a fork from the main Galaxy GutHub code base (via the ProTo project, https://github.com/elixir-no-nels/proto). We are currently in the process of moving away from our current Subversion-based development practices towards a git-based solution, while refactoring and streamlining the installation process. We will provide a full release of the GitHub-based Genomic HyperBrowser source code soon.


The situation the past few years

--------------------------------

In the recent years there have been few major code releases (most of the updates having been silent bug fixes), and little public release of new functionality. This has not been due to little development efforts, but rather the opposite. The number of researchers and master students working on the Genomic HyperBrowser and related projects have never been higher than in the last few years (at present 10 developers, not including master students). The main reasons for the lack of public updates are several:


1. The HyperBrowser team has focused on a major overhaul called the GSuite Tools (see above), which has yet to be finalized and published.

2. The number of developers focusing on new features has increased, while the resources towards quality assurance towards releases has been reduced due to organizational reasons.

3. We have an ambition on making the distribution of the software cleaner and more installation-friendly by moving towards a github-based source code (as explained above). However, the massive number of code branches and features developed in this and related projects (e.g. the ProTo project, https://github.com/elixir-no-nels/proto) have made this a difficult, time-consuming job.


So stay tuned. Major things will happen soon!


For the HyperBrowser team,

Sveinung Gundersen

A new version of the Genomic HyperBrowser has been released!

posted May 13, 2013, 6:22 AM by Sveinung Gundersen

Dear HyperBrowser users,

We are happy to announce the release of version 1.6 of the Genomic HyperBrowser!
This release is a companion to an article recently published in Nucleic Acids
Research, to be included with the 2013 Webserver issue:

   Sandve, Geir K., et al. "The Genomic HyperBrowser: an analysis web server
   for genome-scale data." Nucleic acids research (2013)
   (http://nar.oxfordjournals.org/content/early/2013/04/29/nar.gkt342.full)

In this article, an overview of most of the tools and analysis included with the
Genomic HyperBrowser is presented (for the first time) in table form.


The focus of this release has been on new and revamped tools:

1.  A set of new tools has been added:

   - Visualize track elements relative to anchor regions
   - Create high-resolution map of track distribution along genome
   - Create high-resolution map of multiple track distributions along genome
   - Extract segments where value is greater/less than threshold

2.  A new hypothesis test for 3D-colocalization of genomic elements has been
   added (along with a tool for easier access to the analysis). This hypothesis
   tests checks whether the elements of a genomic track is positioned on the
   DNA (in interphase) closer in 3D than expected by chance. The methodology
   has been published in a separate article:

   Paulsen, Jonas, et al. "Handling realistic assumptions in hypothesis testing
   of 3D co-localization of genomic elements." Nucleic acids research (2013).
   (http://nar.oxfordjournals.org/content/early/2013/04/08/nar.gkt227.full)

   Included with this analysis are readily preprocessed Hi-C datasets for the
   cell lines GM06990, GM12878, hESC, K562, and RWPE1 for human (hg19), in
   addition to cell lines cortex and mEsc for mouse (mm9), all available in
   several resolutions.

3.  The usability of tools has been improved:

   - The structure of the HyperBrowser tools has been rearranged, and renamed
     for easier understanding and use
   - Help text and usage examples (using Galaxy Pages with embedded histories)
     has been added to most tools

4.  Numerous bug fixes and improvements have been made, including the
   fixing of several memory leaks.


We hope you will find the new additions useful. Please let us know if you
experience any problems, need help with your analysis, or have ideas for new
functionality!


For the HyperBrowser team,
Sveinung Gundersen

Version 1.5 of the Genomic HyperBrowser was released sept. 2012

posted May 13, 2013, 6:21 AM by Sveinung Gundersen

The release notes for version 1.5, which was released sept. 2012, is attached below. We forgot to update the announcement page at that time.

--

Dear HyperBrowser users,

We are happy to announce the release of version 1.5 of the Genomic HyperBrowser!

This release has focused mostly on the data handling side of the system, but it also includes much new functionality of
other kinds. New features include:

1.  The recently developed GTrack file format [1] is now fully supported by the Genomic HyperBrowser. In brief, the
    GTrack is a flexible file format developed to handle most kinds of genomic data, previously commonly handled through
    specialized formats like BED, WIG, FASTA and BedGraph. GTrack also handles genomic data describing the 3D
    organizational aspects of DNA, such as data from Hi-C experiments.

    The GTrack specification is easily accessible from within the system. In addition, we have developed a set of 8
    associated web tools, including:

    - Create GTrack file, supporting simple creation of GTrack files from any tabular file with genomic data
    - Validate GTrack file
    - Convert between GTrack/BED/WIG/bedGraph/GFF/FASTA files. This tool also supports conversion between the other file
      formats (not only to/from GTrack)

2.  In the above-mentioned publication [1], we also identified 15 generic track types, characterizing most genomic tracks
    in current use. All 15 track types are now supported by the Genomic HyperBrowser. The data handling backbone of the
    system has thus been thoroughly revised.

3.  The user interface has been revamped:
   
    - The interface is more informative, with better help messages and better user feedback on errors
    - The access to data in the history has been improved, with the extract track and create intensity track tools
      working with data in the history. Also, a single click, perform HyperBrowser analysis link, has been added to
      datasets in the history
    - The design of the web pages has been improved
   
4.  Several improvements have been made with respect to the analyses:

    - The new MCFDR algorithm [2] has been implemented, providing large gains in computational efficiency for hypothesis
      tests making use of Monte Carlo (MC) simulation
    - The explicit setting of the random seed, useful for replicating the exact results of a Monte Carlo run
    - The ability to run analyses only on the regions where the tracks are defined (i.e. the bounding regions of the
      tracks).
    - Better graphs and output of distributions
    - New hypothesis tests (including test of 'Preferential overlap?' on case-control segments), as well as several new
      and improved descriptive statistics

5.  Support for textual specification and modification of runs. For any analysis specified in the GUI, a corresponding
    batch run line can be found in the run description. Such batch lines can be collected, duplicated and modified
    easily as pure text, and can then be pasted into a new Batch run tool that will run all specified batch lines, in
    sequence. Textual modifications allows analyses to be run with several alternative parameter settings, or run on
    alternative tracks.
   
6.  Several new tools, among others:
    - Create categorical BED file
    - Split BED file into sub-regions of a fixed size
   
7.  Numerous small bug fixes and improvements, including fixing a bug that periodically caused erratic behavior in the
    user interface.
   
(Note: The first patch to version 1.5.1 has already been applied, fixing some small bugs in the 1.5 release.)

We would also like to bring two recent publications to your attention [3,4]. These publications are based upon analyses
on genomic tracks related to Multiple Sclerosis. The analyses were run on the Genomic HyperBrowser, and are thus good
examples of how the Genomic HyperBrowser may be efficiently used as the methodological basis for new findings.
       
For the HyperBrowser team,
Sveinung Gundersen

[1] Gundersen S, Kalas M, Abul O, Frigessi A, Hovig E, Sandve GK: Identifying elemental genomic track types and
    representing them uniformly. BMC Bioinformatics 2011, 12:494.

[2] Sandve GK, Ferkingstad E, Nygård S: Sequential Monte Carlo multiple testing. Bioinformatics 2011, 27(23):3235-41.

[3] Disanto G, Sandve GK, Berlanga-Taylor AJ, Morahan JM, Dobson R, Giovannoni G, Ramagopalan SV: Genomic regions
    associated with multiple sclerosis are active in B cells. PLoS One. 2012;7(3):e32281.

[4] Disanto G, Sandve GK, Berlanga-Taylor AJ, Ragnedda G, Morahan JM, Watson CT, Giovannoni G, Ebers GC, Ramagopalan SV:
    Vitamin D receptor binding, chromatin states and association with multiple sclerosis. Hum Mol Genet. 2012,
    21(16):3575-86.

The differential disease regulome

posted Aug 3, 2011, 7:23 AM by Sveinung Gundersen   [ updated Aug 26, 2011, 6:39 AM by Sveinung Gundersen ]

We are proud to present a new paper, extending the Genomic Hyperbrowser with new methodology:

Sandve GK, Gundersen S, Rydbeck H, Glad IK, Holden L, Holden M, Liestol K, Clancy T, Drablos F, Ferkingstad E, Johansen M, Nygaard V, Tostesen E, Frigessi A, Hovig E. The differential disease regulome. BMC Genomics. 2011 Jul 7;12(1):353. PubMed PMID: 21736759.
 
The paper describes a method for creating large-scale maps, giving global overview of the relations between two sets of tracks. The Differential disease regulome is the main example of the methodology, mapping Transcription Factors against all human diseases, leading to a global map of around 900 000 TF-disease pairs.

The Differential disease regulome and other regulomes can be viewed in its main version from the HyperBrowser web page. The regulomes are found either under "Regulomes" in the tools panel, or from a link on the welcome page. The relations between diseases and TFs can be explored using interactive Google Maps technology. Background information on the genes that make up the relations can be fetched using the marker and cluster tools. You can also create your own regulomes by selecting subsets of the transcription factors and diseases, using the regulome tools from the tools panel.

For the HyperBrowser team,
Sveinung Gundersen

Streamlined HyperBrowser installation

posted May 13, 2011, 4:01 AM by Geir Kjetil Sandve

Dear users,

For those interested in making a local install of the Genomic HyperBrowser, the installation process has now been streamlined.

After having installed a local instance of Galaxy, and a few other dependencies like Python itself, installation of HyperBrowser should now in principle just amount to running a single setup.py with appropriate parameters.

From the front page of the HyperBrowser (http://hyperbrowser.uio.no), you can follow the link near the bottom of the page (where it says "The source code is available for download from this page."). There you will find how to obtain our latest version from a SVN repository. After checking out the code, you will in the trunk find a file INSTALL.TXT with brief installation instructions, as well as setup.py to start installation.

We are of course happy to answer any questions you might have.

Regards,
Geir Kjetil Sandve
The HyperBrowser team

The Genomic HyperBrowser: Version 1.0

posted Feb 23, 2011, 7:07 AM by Sveinung Gundersen

Dear users,

We are happy to announce that version 1.0 of The Genomic HyperBrowser has been released. This release has mainly focused on the addition of new tools and functionality.:

Export / import tools:
- Extract track: Added the functionality of extracting each region to a separate file
- Extract IDs of intersecting genes: New tool that finds the IDs of genes intersecting with a track.

Create tracks:
- Create track from DNA sequence: New tool that creates a function track based on a custom expression applied on a sliding windows across the genome.
- Create smooting or density track: New tool that creates a density distribution or smoothed version of any track, based on a sliding window across the genome.

Nmer analysis:
- Analyze nmer occurrences: New tool used to analyze occurrences of any nmer (oligonucleotide) along the genome, either in itself or in relation to other genomic tracks.
- Inspect nmer frequency variation: New tool that inspect the frequency variation of occurrences of any nmer in a specified region along the genome (or genome-wide).

Plot-based analysis:
- Scatter plot of track relation: New interface to already existing functionality for creating scatter plots of the relation between two tracks in local bins along the genome.
- Plot of progression pattern: New interface to already existing functionality for creating plots of track progression patterns that are common across different bins.

Version 1.0 also includes some bug fixes and under-the-hood improvements.

We hope that you find the new additions interesting. If you have any questions, requests or comments of any kind, please feel free to use our discussion forum, available from the Help menu, or send an email to hyperbrowser-requests@usit.uio.no. We are happy to respond.

Regards
Sveinung Gundersen
The HyperBrowser Team

Scheduled downtime

posted Feb 2, 2011, 5:32 AM by Sveinung Gundersen

Notice to users of the Genomic HyperBrowser,

The server hosting hyperbrowser.uio.no will have a scheduled downtime Monday, 7 february, due to network maintenance. The Genomic HyperBrowser will be down from 0900 to 1600 CET. We apologize for any inconvenience this may cause.

Regards,
Sveinung Gundersen
The HyperBrowser team

First article published!

posted Dec 27, 2010, 2:38 AM by Sveinung Gundersen   [ updated Aug 3, 2011, 8:13 AM ]

The first article on The Genomic HyperBrowser was published in Genome Biology, Thursday, 23 December 2010:

Sandve GK, Gundersen S, Rydbeck H, Glad I, Holden L, Holden M, Liestol K, Clancy T, Ferkingstad E, Johansen M, Nygaard V, Tostesen E, Frigessi A, Hovig E. The Genomic HyperBrowser: inferential genomics at the sequence level. Genome Biol. 2010 Dec 23;11(12):R121. [Epub ahead of print] PubMed PMID: 21182759.

This was a welcome holiday gift!

If you have any comments, questions or ideas, please drop us an email, or contribute to the discussion forum.

For the HyperBrowser team,
Sveinung Gundersen

1-8 of 8