0.3.0 - 2018-04-01
==================
* IMPORTANT: Default number of threads in `wellington_footprints.py` now set to 1 - set to 0 to use all cores
* BUGFIX: Fix Cython code for latest Cython
* BUGFIX: Logic fixes to underlying Wellington C code (@dpryan79)
* BUGFIX: Shebangs added to python scripts (@dpryan79)
* BUGFIX: Fix ability to compare `GenomicInterval` objects in python3 (@dpryan79)

0.2.6 - 2018-01-27
==================
* ENHANCEMENT: License changed to MIT
* BUGFIX: Scripts should now all work for Python 3
* BUGFIX: Empty lines in BED files are now ignored
* BUGFIX: Fix issue with passing values to the "-pv" argument on `wellington_footprints.py`

0.2.5 - 2016-07-23
==================
* IMPORTANT: Python 2.6 and 3.2 support dropped. 2.7, 3.3, 3.4, and 3.5 are where it's at.
* BUG: pyDNase now ignores unmapped reads in the BAM input file. NOTE: Will not ignore secondary alignments in paired reads. Filter these out beforehand if you need this behaviour for the timebeing.
* BUG: Fix `dnase_bias_estimator.py` script arguments.
* BUG: Fix `frd_limit` flag not being able to be set to 0 in `wellington_footprints.py` and `wellington_bootstrap.py`.

0.2.4 - 2016-05-30
==================
* BUG: Update author’s contact details

0.2.3 - 2016-01-17
==================
* BUG: Fix error in ``wellington_bootstrap.py`` which caused no output to be written to disk (thanks to Duy Pham for reporting)

0.2.2 - 2015-12-26
==================
* BUG: Fix error in ``wellington_bootstrap.py``

0.2.1 - 2015-12-25
==================
* ENHANCEMENT: ATAC-seq mode for ``wellington_bootstrap.py``

0.2.0 - 2015-12-18
==================
* FEATURE: Differential Footprinting between treatment and control (``wellington_bootstrap.py``)
* FEATURE: Estimate 6-mer bias cleavage (``dnase_bias_estimator.py``)
* FEATURE: ATAC-seq compatability! (pass -A to most of the scripts to enable ATAC-seq read shifting for the transposase staggered integration)
* FEATURE: Preliminary DNase I cleavage bias correction in ``dnase_average_profile.py`` and ``dnase_to_javatreeview.py`` (see `here <http://pythonhosted.org/pyDNase/biascorrecting.html>`__ for full documentation)
* FEATURE: Calculate ∆DHS `(He et al. 2012) <http://www.ncbi.nlm.nih.gov/pubmed/22508765>`__ scores between two DNase-seq runs (``dnase_ddhs_scorer.py``) 
* FEATURE: Annotate BED file number of cuts in each region (``dnase_cut_counter.py``)
* ENHANCEMENT: DNase-seq footprinting tutorial :)
* ENHANCEMENT: In the interests of performance, ``BAMHandler`` no longer returns NumPy arrays, just lists.
* ENHANCEMENT: ``wellington_footprints.py`` is now multithreaded (woo) - performance roughly scales linearly with number of CPUs.
* ENHANCEMENT: The ``Footprinting`` module has been refactored to allow for multithreading - those using the Wellington API directly take note of the changes.
* ENHANCEMENT: Everything is faster! More cythonised code.

0.1.7 - 2014-09-03
==================
* BUG: Fixed bug in dnase_to_javatreeview.py that prevented caching from being disabled

0.1.6 - 2014-03-07
==================
* BUG: Fixed FDR calculation that was only performing 1 randomisation! Many thanks to @arjanvandervelde

0.1.5 - 2014-01-31
==================
* ENHANCEMENT: Removed SciPy requirement
* ENHANCEMENT: Added Travis-CI automated builds for 2.6, 2.7 and 3.3 and coveralls
* BUG: Fixed WIG output for UCSC compatibility

0.1.4 - 2014-01-05
==================
* BUG: Fix dodgy deployment

0.1.3 - 2014-01-05
==================
* BUG: Fixed Python 2.6 Compatibility (again!)

0.1.2 - 2013-12-09
==================
* BUG: Fix issue where BED intervals with chromosome names not starting with "c" were silently being ignored (reported by Aaron Hardin)
* BUG: Fix clint dependency issue (no longer requires custom version of clint)
* BUG: Fix spelling error in CHANGES

0.1.1 - 2013-12-05
==================
* BUG: Misc. small bug fixes
* ENHANCEMENT: Fixed Python 2.6 Compatibility
* ENHANCEMENT: Added JSON export script

0.1.0 - 2013-09-01
==================
* Initial Release