CHANGELOG

v0.4.0
- Removed proteins without any PSMs below the FDR threshold from protein groups output. Setting the new `--keep_all_proteins` flag reverts to the old behavior.
- Added support for supplying multiple evidence files
- Added support for mokapot format and allow users to specify methods with custom toml files
- Add `--fasta_contains_decoys` flag and functionality
- Update llvmlite (#4)

v0.3.5
- Added rescoring input (e.g. Prosit results) to GUI

v0.3.4
- Added percolator input to GUI

v0.3.3
- Add option for multiple evidence files and selecting digestion parameters in GUI

v0.3.2
- Fix mokapot decoy filename
- Propagate argparse errors from pipeline scripts to logger

v0.3.1
- Fixed missing methods folder in GUI

v0.3.0
- Added make rules to reproduce manuscript figures
- Moved methods to toml files
- Update picking strategy to use leading proteins
- Add max_workers option for mokapot
- Add support for multiple threads in LFQ computation
- Fix non-functioning train_fdr option for mokapot

v0.2.7
- Added maxtasksperchild for better memory management in LFQ

v0.2.6
- Upgrade minimum Python version to 3.8 for new scipy version
- Reduced LFQ memory by 80% and speed up by 30%

v0.2.5
- Undo switching order of log and median operation in LFQ as this led to wrong results

v0.2.4
- Speed up LFQ by using Bottleneck package for nanmedian

v0.2.3
- Added unit and performance tests to repo
- Allow multithreading for LFQ quant

v0.2.2
- Upgrade minimum Python version to 3.7 to remove dataclasses dependency

v0.2.1
- Change requirements to inequalities for packages with 0.x version to allow better dependency resolving

v0.2.0
- Major improvements to GUI and removed unnecessary libraries to reduce file size

v0.1.0
- Initial release