Version 5.0
===========

Version 5.0.0
-------------

This version adds the following features, outlined below. Please note that
pVACtools 5.0 is not guaranteed to be backwards-compatible and certain changes
could break old workflows.

New Tools
_________

This release adds a new tool, pVACsplice, for prediction neoantigens from
splice sites. Please see the :ref:`full tool documentation <pvacsplice>` for more information.
by @mrichters in https://github.com/griffithlab/pVACtools/pull/911

New Features
____________

- This release refactors the pVACvector graph building algorithm in order to increase the probability
  for finding a solution and reduce the number of iterations needed before a solution is found. Please
  see the PR describtion for the full details. by @susannasiebert in https://github.com/griffithlab/pVACtools/pull/1163
- Add a new ``--aggregate-inclusion-count-limit`` parameter to set the maximum number of epitopes
  to include in the metrics.json detailed data for variants that have a large number of candidate
  neoantigens (e.g., frameshifts). by @susannasiebert in https://github.com/griffithlab/pVACtools/pull/1147
- Add a new ``--biotypes`` parameter which defines a list of transcript consequence biotypes that the
  predictions from pVACseq and pVACsplice should be limited to. by @mrichters in https://github.com/griffithlab/pVACtools/pull/911
- Add support for additional alleles that weren't previously supported, includings ones for dog,
  mouse, and MHC class II. by @susannasiebert in https://github.com/griffithlab/pVACtools/pull/1148

Bugfixes
________

- This relase fixes a bug with the ``--agregate-inclusion-binding-threshold`` which would not be used if
  the ``--allele-specific-binding-thresholds`` flag was set. by @susannasiebert in https://github.com/griffithlab/pVACtools/pull/1147
- The pVACview percentile plots have been updated to include percentiles from elution and immunogenicity
  algorithms. by @susannasiebert in https://github.com/griffithlab/pVACtools/pull/1149
- This release fixes a bug where the incorrect neoantigen fasta entry may be used for the reference proteome
  search if there were multiple variants or alt alleles located at the same genomic position. by @susannasiebert in https://github.com/griffithlab/pVACtools/pull/1153
- Add additional trailing amino acids for frameshift insertions when creating fasta in order to capture a
  matched wildtype entry in large repetitive regions. by @susannasiebert in https://github.com/griffithlab/pVACtools/pull/1155

Version 5.0.1
-------------

This is a bugfix release. It fixes the following problem(s):

- The ``--aggregate-inclusion-count-limit`` was not being passed along to the
  aggregate report module in the pVACsplice pipeline correctly. This release
  fixes that issue. by @susannasiebert in https://github.com/griffithlab/pVACtools/pull/1167