Skip to content

Latest commit

 

History

History
146 lines (119 loc) · 6.15 KB

CHANGELOG.rst

File metadata and controls

146 lines (119 loc) · 6.15 KB

Changelog

List of notable changes to master release of this code should be documented here. Hopefully I can keep up with this!

[1.3.030118] - 2018-03-01

  • Added full set of documentation, including a html version (to be hosted on readthedocs in the near future and as a part of the upcoming Github repository), as well as a PDF and Man pages version.
  • Minor code tweaks to fix a few bugs here and there.
  • Code cleanup to help better adhere to standard PEP requirements.

[1.2.021418] - 2018-02-14

  • Fixed some issues with setup.py. Was not properly configuring every component, especially with regard to setting up requirements.
  • Changed location of root_dir definition from postinstaller to __init__. Was having problems with setup.py when not defined in __init__.
  • Minor code refactoring and cleanup.

[1.0.021418] - 2018-02-14

  • Removed failed biopsies from output when calling MatchData.get_bsn().
  • Fixed PSN mapping if one does not use PSN string in name.
  • Fixed bug with file handle creation for output.

[0.22.012218] - 2018-01-22

  • Added MatchData.get_ihc_results(). This method will return IHC data results based on an input patient identifier.
  • Added back variant measurements (VAF, CN, Read Counts) to MatchData.find_variant_frequency(). Not sure anymore why I dropped it in the first place!

[0.21.011818] - 2018-01-18

  • Added "outside confirmation" to list of things to skip in processing / output.
  • Fixed all method calls to new versions in the match_data module.
  • Fixed MatchData.find_variant_frequency() for the following:
    • Better handle outside assay results; we now skip them!
    • Changed order of counts in output.
    • Output a patient based conter and biopsy based counter. This is to help discern between progression biopsy cases.
  • Added MEDRA code to CTEP Disease identifier lookup method.
  • Completely overhauled MatchData.get_disease_summary(). Method now:
    • Handles queries better and more efficiently.
    • Now outputs MEDRA code information.
    • Output format cleaned up
  • Added ID and counts output for MatchData.get_biopsy_summary().
  • Fixed bug with PSN string stripping in MatchData.find_variant_frequency()

[0.20.0] - 2017-10-27

  • Updated and wrote package tests.
  • Added warning for no matches when looking at non-outside assay cases. Not ideal, but as the backend of MATCHBox becomes less and less reliable due to this really dumb Outside Assays thing, I need better messages than just generic Python stack trace errors.
  • Added Python 3 compatibility. At the moment don't want to force python3 on to users (some may find it difficult to install), but I want this to ultimately be totally run under Python 3. Should only need some small tweaks once this goes full python 3.
  • Change main MatchData api constuctor to use sys_default JSON file of data rather than doing a live query. This will really speed things up, and now that MATCHBox data is becoming more and more stable as incoming data slows down, this is perfectly fine.
  • Added MatchData.get_patient_by_disease() method to, well, get patient by disease!

[0.19.1] - 2017-10-03

  • Updated tests.
  • Some minor tweaks to new data structure added.
  • Complete overhaul on MatchData.get_histology(). Working much more smoothly now!
  • Changed some internal private methods to better handle new data formats and add better functionality to internal queries.

[0.18.0] - 2017-09-28

  • Completely overhauled the data structure so that we can now better handle multiple biopsies and NGS results (i.e. MSNs) for the same patient. Cases where there were progression biopsies and the like were breaking methods.

  • Refactored key methods like:

    • MatchData.get_psn()
    • MatchData.get_bsn()
    • MatchData.get_msn()
    • MatchData.variant_frequencey()
    • MatchData.get_biopsy_summary()

    Not a complete list, but good representation. These methods now use the new data structure to better represent results, and in cases like MSN return, output data as lists since we know we can have multiple MSNs / PSN.

[0.17.0] - 2017-09-22

  • Updated tests.
  • Fixed some status reporting bugs, including a hardcoded (and not ideal!) fix for 4 odd compassionate care cases for Arm P. Still need to remove soem debugging code, but will leave that until 100% sure this is all correct.
  • Added MatchData.get_patient_meta() a method to just return raw chunks of JSON file data for a patient based on an input key. This is mainly for debugging and devlopment and may end up being privatized in release versions.
  • Code clean up and finishing patient level TA data in method calls.

[0.16.0] - 2017-09-19

  • Created a new treatment_arms module with a TreatmentArms() class that will allow for adding treatment arm data and aMOI rules assignment. As a part of arm compilation work, needed to know what was an aMOI and what was not!
  • Started project documentation by adding better python docstrings to code (really just updating and making better what I had already written), and adding things in like this CHANGELOG. Will begin experimenting with the Sphinx documentation system soon to figure out how to make really nice release docs as I'm anticipating many people ultimately using these tools. Well...hopefully!