Version 0.5.0

0.5.0 - 2023-07-18

Fixed

• Properly handling Ns in the MSA, and in the denovo sequences (see PR #60
  and PR #61 for more details);

Changed

• scikit-learn, numpy and pytest dependencies updated;
• The KMeans algorithm used is now elkan;

Version 0.4.0

0.4.0 - 02/12/2022

Added

• make_prg update command, that updates PRGs without requiring to rebuild MSAs and the PRG itself from scratch;
• Trace (-vv) logging level, to track make_prg behaviour (intended for developers only);
• Multithreading support (-t parameter);
• A sample example;
• 365 new tests (from 116 to 481 total tests), with test coverage >99% in non-argument parsing code;
• Precompiled binary;

Changed

• make_prg from_msa: input can now be a single file or a directory. If it is a single file,
  a <prefix>.prg.bin, a <prefix>.prg.fa, a <prefix>.prg.gfa and a <prefix>.update_DS.zip
  files are created. If it is a directory, all files in the directory are scanned and the same
  execution for a single file is done for each input file found. The output files are a collection of the single-input
  execution: a <prefix>.prg.bin.zip file will contain a collection of .prg.bin files, similar to
  <prefix>.prg.gfa.zip and <prefix>.update_DS.zip; <prefix>.prg.fa will be a multi fasta;

• Other make_prg from_msa CLI changes (please run make_prg from_msa -h for a full description of the new parameters):

  • Parameters removed: --prg_name, --seqid, --no_overwrite;
  • Parameters added: -s, --suffix, -F, --force, -t, --threads, -g, --output-graphs;
  • Parameters changed: Replaced --outdir by --output_prefix;

• The recursive clustering and collapse algorithm is now explicitly represented as a tree with internal data
  structures that remember the multiple sequence subalignment at any point of the recursion, as well as several other
  internal data, allowing the serialization and deserialization of the recursion tree at any point. Thus updates can
  be done avoiding any recomputation by firstly saving the state of the recursion tree to disk, and then loading this
  recursion tree, adding denovo sequences to some specific nodes, and triggering recomputation of the modified nodes.
  Any preorder traversal of the recursion tree yields the same order of recursive calls of the previous algorithm,
  thus allowing us to translate the algorithms in the previous version as preorder traversals with custom visit
  operations.

• Moved from setup.py to pyproject.toml

Fixed

• Several minor bugs;
• Heavy refactoring of almost the whole codebase;

Removed

• Dropped support for python 3.7, supported python versions are: 3.8, 3.9, 3.10, 3.11.
• Dropped support for Mac OS X;

Version 0.2.0

New command-line

• Added:

  • -S, --seqid option to name the PRG sequence, which by default uses the file name.
  • -N shortcut for max nesting
  • -L shortcut for min match length
  • --log to enable specifying log file should go to path. Default behaviour is now that
    log goes to stderr by default
  • -O, --output-type option to specify what output files are required. Defaults to
    all

• Removed:

  • summary file

• Changed:

  • --prefix CLI parameter of from_msa subcommand removed in favor of CLI parameters --outdir
    and --prg_name, with sensible defaults (current working directory and MSA file name stem respectively).
    This allows finer control over where to place output files.

Output files

• Output files:
  • No longer contain 'max_nesting' and 'min_match_length' in their names; these appear in the log files,
    and in the .prg fasta header.
  • .bin file now stores even integer markers at site ends; this is the format used by gramtools.
  • Summary file not written by default

Bug fixes

• Bugfix ([#27])- Part 1: clustering termination was not properly detected, causing spurious
  site production
• Bugfix ([#27])- Part 2: added criteria for not clustering, to reduce construction of
  ambiguous graphs

Version 0.1.1

0.1.1 - 2021-01-27

Added

• Dockerfile
• -V option to get version

Changed

• A test that was clustering all unique 5-mers was reduced to all 4-mers as the memory
  usage of all 5-mers was causing a segfault when trying to run the tests during the
  docker image build.

Removed

• Singularity file as it is redundant with the new Dockerfile (that will be hosted on
  quay.io)
• scipy dependency. We never actually explicitly use scipy.

0.1.0

Added

• This CHANGELOG file to hopefully serve as an evolving example of a standardized open
  source project CHANGELOG.