Version 0.5.0 - Parallel processing

New features:

• support processing multiple location in parallel (--n-cores)
• support other input than tally of SVN mutation, e.g., use tables
  produced by COJAC.
• option to specify a column to use as 'names' for the entries in tally table.
  By default, if 'pos' and 'base' exist a column 'mutations' will be created
  and used as name.
  Bootstraping needs such names for the entries.

Results altering change:

• The random seed initialisation has changed since 0.4.x
  Results between this version v0.5.0 and v0.4.x aren't bit-perfect
  Seed is re-initialised per locataion so parallel and sequencial runs
  are identical.
  Each location gets a different seed.

v0.4.1 filters bugfix

bugfix:

• empty --filter=""
• fixe quoting regex

v0.4.0 - more general

Features:

• configurable filters to remove problematic mutations from the table
• supports runing LolliPop on other generic type of timeseries,
  not necessarily single point mutations

Bugfix:

• ruamel >= 0.15

0.3.0 - V-pipe integration

• Converted to a proper python package with poetry and click
  • Command format has changed and uses subcommands
• Integrated into V-pipe
• Outputs JSON for dashboard uploads
• Tools for generating input to deconvolution (tallymut)
• nodate mode (for benchmarks w/o timeline)
• Presets and examples
• Automatic test (does first/fastest curve from preprint)
• Partial support for aarch64
  • Optionnal BCBio still missing on bioconda

Bugfixes:

• half-height curves

  if a variant in the var_date interval and one not being considered
  are very close to each other, there is a risk that bootstraping
  will attribute only half to the variant and half to "undetermined"

Publication:

• Notebook graphs can now be replicated from command line

0.1 - prerelease for Recomb 2023

This corresponds to the version used in the notebooks for the article submitted at Recomb 2023