Skip to content

Latest commit

 

History

History
139 lines (117 loc) · 5.98 KB

CHANGELOG.md

File metadata and controls

139 lines (117 loc) · 5.98 KB

2022-03-14 - v3.2 (pending)

  1. [bugfix] -- ALL.reformatGeneFamilies.py doesn't need the ancestral species to have genes
  2. [bugfix] -- NHX trees can now miss internal labels, as AGORA can generate names by itself.
  3. [bugfix] -- Species names in the gene trees are not transformed by AGORA anymore, and simply have to match the names in the species tree.
  4. [new] -- misc.compareGenomes.py now defaults to colouring the chromosomes according to their size

2022-02-05 - v3.1

  1. [new] -- The agora-*.py scripts now accept existing orthology groups as an alternative to gene trees.
  2. [new] -- New -compress option to change the compression algorithm (or skip compression).

2021-03-12 - v3.0

  1. [major change] -- Migrated to Python 3.
  2. [change] -- Settled on using "blocks" for all reconstructions: no contigs / scaffolds any more.
  3. [change] -- Removed the scripts used to convert to our own tree format since AGORA now natively supports Newick and NHX.
  4. [change] -- buildSynteny.integr-scaffolds.py now outputs all the support scores, incl. the ones from the input blocks.
  5. [new] -- Up-to-date help messages for all scripts.
  6. [new] -- Option to sort the output genomes by decreasing block size.
  7. [new] -- Workflow steps are now skipped if they have already been run (plus an option to force rerunning an entire workflow).
  8. [new] -- Dropped the dependency on joblib (Python's multiprocessing now works well enough for our needs).
  9. [bugfix] -- Translated some French comments.

2021-02-07 - v2.1

  1. [change] -- Changed the parameters of gene-families filtering in the Vertebrates workflow to the values used for Genomicus production.
  2. [new] -- Support for macOS (Catalina).
  3. [new] -- Added the ability to do the second reconstruction pass (scaffolds) in a multi-integration fashion in the library myAgoraWorkflow.
  4. [new] -- New preset to run reconstructions for Plants (agora-plants.py).
  5. [new] -- New script to test all reconstruction parameters and automatically select the best. It should apply to any clade (agora-generic.py).

2020-12-07 - v2.0

  1. [major change] -- Renamed all scripts and most options to match the upcoming publication.

2020-10-01 - v1.5

  1. [bugfix] -- Added the missing +onlySingletons option to the buildSynteny.integr-extend.py call in agora2.py.
  2. [change] -- New -extantSpeciesFilter parameter to control which extant species can be used to run pairwise comparisons (buildSynteny.pairwise-conservedPairs.py and buildSynteny.integr-groups.py). The option is exposed in all workflow scripts. In buildSynteny.integr-groups.py it replaces the third positional command-line argument.
  3. [change] -- Removed the multithreading option of buildSynteny.integr-copy.py as it does not give any benefits.
  4. [change] -- Run the workflow steps sequentially by default in the agora scripts, with an option to enable their parallelisation.
  5. [new] -- Print memory usage stats when running workflows.
  6. [new] -- Decreased the memory usage of ALL.filterGeneFamilies-size.py, buildSynteny.pairwise-conservedPairs.py, and buildSynteny.integr-groups.py.

2020-06-25 - v1.4

  1. [bugfix] -- Avoid ALL.extractGeneFamilies.py crashing because of a ValueError in certain conditions.
  2. [bugfix] -- Some mitochondrial genomes were excluded from the reconstructions.
  3. [bugfix] -- buildSynteny.integr-refine.py doesn't crash with a ZeroDivisionError when there isn't a previous score to compare against.
  4. [change] -- buildSynteny.integr-refine.py and buildSynteny.integr-groups.py can now run across multiple cores.
  5. [change] -- The path evaluation parameter of buildSynteny.integr-refine.py now defaults to the one used in the paper.
  6. [change] -- Removed two Python dependencies: numpy and enum.
  7. [change] -- Unary nodes are now allowed in the species-tree.
  8. [new] -- New library (myAgoraWorkflow) to simplify the creation of workflows around AGORA scripts.
  9. [new] -- Print CPU usage stats when running workflows.
  10. [new] -- New scripts (agora1.py and agora2.py) to run the default workflows.

2020-05-13 - v1.3

  1. [change] -- Moved the LibsDyoGen inside scripts/utils/
  2. [change] -- buildSynteny.integr-groups.py now accepts _ as a value for the usedSpecies parameter and interprets it as the same value as the target parameter.
  3. [change] -- ALL.extractGeneFamilies.py now outputs trees in the same format as the input trees

2020-05-10 - v1.2

  1. [bugfix] -- agora.py now makes the script it launches follow the nbThreads option.
  2. [bugfix] -- Can now have multiple ancGenes size instructions in the configuration file.
  3. [bugfix] -- agora.py now applies correct dependencies for the pairwise comparisons
  4. [change] -- Moved and renamed the scripts that convert file formats.
  5. [new] -- All scripts now natively support Newick species tree and NHX gene trees.
  6. [new] -- New publish instruction to convert the ancestral genomes from the diags format to ancGenomes.
  7. [new] -- agora.py can now generate the first set of ancGenes (all)

2020-04-27 - v1.1

  1. [bugfix] -- agora.py was not detecting task failures.
  2. [bugfix] -- agora.py was always creating an ancGenes-filter task even when it doesn't need one.
  3. [change] -- Added a -LOG.ancGraph option to all the buildSynteny.integr scripts (instead of hardcoded paths). The integrOutput option of the configuration files now refer to this per-ancestor log file, rather than the standard output of the scripts (which is now empty).
  4. [change] -- Relative paths for input genes and species-tree are now evaluated from the location of the configuration file.
  5. [change] -- Following a rewrite of the HowTo, the structure of the output directories as defined in the configuration files has changed:
    • diags/pairwise → pairwise,
    • diags/integr → integrDiags.

2016-09-15 - v1.0

Initial release