BrowseVCF is a web-based application and workflow to quickly prioritise disease-causative variants in VCF files.

July 2020 update: BrowseVCF is not officially supported
While we're happy BrowseVCF has proven helpful to many users, we unfortunately no longer have time to maintain it and to keep it up to date with changing VCF format specifications. If you encounter any problems, please check among the known (closed) issues section. We may someday try to update the code and fix the known bugs (no promises though!), but you're very welcome to submit patches and pull requests to fix outstanding issues.

1. Requirements and Installation
2. For Windows users
3. For GNU Linux users
4. For Mac OS users
5. Usage as standalone web application
6. Usage as command line tool
7. Contact and Contribute
8. License

Before proceeding with the installation, please make sure you have enough disk space! For example, the files generated by BrowseVCF during the analysis of an annotated whole-genome sample (~500Mb when compressed) can take ~10Gb.

Download the zip file (browseVCF_win7_vX.X.zip) from the release page, unzip it in a path that does not contain spaces, and double-click on launcher-windows.bat in the web directory.

Download the GNU/Linux-specific .tar.gz file on the release page, unzip it in a path that does not contain spaces, and run launcher-gnu.sh from a Terminal within the web directory. This version is shipped with a stripped-down Python v2.7 compiled on GNU/Linux, along with pre-compiled wormtable and BerkeleyDB modules. Tested on Ubuntu version 14.04.

Alternatively, to compile from source, please follow the instructions below.

1. The latest versions of CentOS, Fedora, Redhat and Ubuntu come with Python 2.7 out of the box. If it's not installed, download Python from https://www.python.org/.

2. Install pip
   wget https://bootstrap.pypa.io/get-pip.py
python get-pip.py

3. Install dependencies
   sudo pip install psutil
sudo pip install cherrypy
sudo pip install cherrypy-cgiserver

4. Install Berkeley DB
   sudo apt-get install libdb-dev (Ubuntu/Debian) or yum install libdb-devel (Red Hat/Fedora)

5. Install Wormtable
   sudo apt-get install python-dev (Ubuntu/Debian) or yum install python-devel (Red Hat/Fedora)
   sudo pip install wormtable

6. Download BrowseVCF (substitute X.X with latest version)
   wget https://github.com/BSGOxford/BrowseVCF/archive/vX.X.tar.gz
tar -xvf vX.X.tar.gz

Download the OSX-specific .tar.gz file on the release page, unzip it in a path that does not contain spaces, and run launcher-osx.sh from a Terminal within the web directory. This version is shipped with a stripped-down Python v2.7 compiled on OSX, along with pre-compiled wormtable and BerkeleyDB modules. Tested on OSX El Capitan.

Alternatively, to compile from source, please follow the instructions below.

1. The latest versions of Mac OS X come with Python 2.7 out of the box. If it's not installed, download Python from https://www.python.org/.

2. Install pip
   wget https://bootstrap.pypa.io/get-pip.py
python get-pip.py

3. Install dependencies
   sudo pip install psutil
sudo pip install cherrypy
sudo pip install cherrypy-cgiserver

4. Install Berkeley DB
   sudo port install db53

5. Install Wormtable
   sudo pip install wormtable
CFLAGS=-I/opt/local/include/db53 LDFLAGS=-L/opt/local/lib/db53/ python setup.py build
sudo python setup.py install

6. Download BrowseVCF (substitute X.X with latest version)
   wget https://github.com/BSGOxford/BrowseVCF/archive/vX.X.tar.gz
tar -xvf vX.X.tar.gz

BrowseVCF used as web application is composed of four steps:

1. Upload and pre-process your input .vcf or .vcf.gz file
2. Create indexes for one or more annotation fields of interest
3. Filter variants according to different criteria/fields/cutoffs
4. Export results and query history

A more detailed tutorial is provided as PDF from the release page, together with a VCF sample file. The PDF tutorial shows how to apply the different filters on the VCF sample file with the most frequenctly used queries.

The folder scripts inside web contains the key set of Python scripts that perform the same actions of the web application:

• script01_preprocess.py -> Essential. Must be executed as first.
  It preprocess the input vcf file in order to be compatible with wormtable.
• script02_convert_to_wt.py -> Essential. Must be executed as second.
  It creates the indexes that will be used to query the annotation fields of interest.
• script03_filter_field.py -> Discretionary. Corresponds to filter A of the web application.
  Performs queries on a given field of interest any of the following operators: 'greater_than', 'less_than', 'equal_to', 'contains_keyword', 'is_absent', 'is_present'.
• script04_select_genotype.py -> Discretionary. Corresponds to filter B of the web application.
  Filters variants based on their genotype in one or more samples.
• script05_region_of_interest.py -> Discretionary. Corresponds to filter C of the web application.
  Keeps only those variants located within the input region of interest.
• script06_get_type.py -> Discretionary. Corresponds to filter D of the web application.
  Allows to select variants of a given type ('SNPs', 'InDels', 'MNPs').
• script07_use_gene_list.py -> Discretionary. Corresponds to filter E of the web application.
  Keeps only variants annotated with one of the gene names/IDs provided as input.

Scripts 01 and 02 must be run in this specific order before being able to run any of the other scripts (03, 04, 05, 06, 07).

To see the list of required and optional parameters of any script, simply write:

$ python script_name.py --help

Want to contribute? Great! Simply report a new issue on GitHub or write an email to:

• Silvia Salatino: silvia (AT) well (DOT) ox (DOT) ac (DOT) uk
• Varun Ramraj: varun (AT) well (DOT) ox (DOT) ac (DOT) uk

BrowseVCF is available under the GPL v3 license.