Skip to content

Latest commit

 

History

History
49 lines (39 loc) · 2.25 KB

CHANGES.md

File metadata and controls

49 lines (39 loc) · 2.25 KB

CHANGES.md - ncfp

v0.2.1a1

  • add ruff.toml configuration
  • move from CircleCI to GitHub Actions for CI
  • if all other attempts to match the linked CDS fail, and there is only one CDS in the retrieved record, use that CDS as the candidate sequence (#46)
  • fix badges (Zenodo/PyPI) in README.md
  • make Stockholm domains work with NCBI sequence inputs (#27)
  • update requirements to avoid deprecated requests-cache (#26)
  • add --use-protein-ids option to use the same protein sequence seqID for the protein and the downloaded nucleotide sequence (#21, Emma Hobbs' first contribution)
  • updated code to use pathlib internally instead of os.path operations
  • fixed loglevel bug with logging to file (#25)
  • update progress bar labels with step numbering (#24)
  • update UniProt queries to use the new API (#35)
  • add --allow_alternative_start_codon option (#34)
  • warn, rather than halt, when a suspected IPG input sequence is found (#34)
  • add extra information when any other than one coding sequence is found for an input protein (#34)
  • add last ditch attempt to identify proteins in an NCBI nucleotide file using the gene name
  • use the UniProt accession number if the GN field is not specific enough
  • when UniProt records point to NCBI gene records, obtain the nucleotide accession from the gene record, and protein ID from the RefSeq cross-reference
  • use the UniProt ORF gene name field for identifying CDS in larger records, to avoid issues with ambiguous GN fields
  • adapt to bioservices 1.12 change that modifies UniProt return string format

v0.2.0

  • bump version and add tag for release

v0.2.0-a1

  • Update copyright notices
  • Convert parsers to use pathlib
  • Add requirements file for development tools
  • Convert tests to use pytest, not nose
  • Revise logging usage
  • Change continuous integration from TravisCI to CircleCI
  • Guess sequence origin rather than asking user to provide at CLI (allows mixed origin files)
  • Update CLI parser and docs to reflect new sequence origin guessing

v0.1.1

  • Add installation instructions and other improvements to documentation
  • Tidied codebase in some places (removing print statements, unused functions, etc.).
  • Add CLI tests.
  • Correct ncfp program name in help/usage text.

v0.1.0

*- First release of ncfp