Skip to content

v1.10 - Novel in & novel out

Compare
Choose a tag to compare
@oschwengers oschwengers released this 14 Nov 15:43
· 6 commits to main since this release

This is the tenth minor release (v1.10) introducing user-provided HMMs, output file recovery, feature inference scores, and various improvements.

Compatible database scheme version: 5

Important

Since v1.10.0, Bakta requires Python >=3.9, uses pyCirclize instead of Circos, and discarded support of DeepSig!

An important decision had to be made for this release regarding supported Python versions, external dependencies and features impacted by this. Both Circos and DeepSig seem to be out of support for a long time. Hence, Circos was replaced by pyCirclize, a pure-Python actively-maintained library, enabling a couple of new features. As a result, Bakta's Python dependency had to be bumped to >=3.9, thus unfortunately loosing compatibility with DeepSig. Dropping an existing feature feels odd and wrong, but as a developer, sticking to unmaintained external software for too long constantly increases your daily pain level and slows down the project as a whole. This hasn't been an easy decision, but a necessary. So, If you depend on the detection of signaling peptides, please keep using Bakta <=v1.9.4.

New features

Improvements

  • Support translation table 25: #323 (Thanks @ndombrowski)
  • Export CRISPR and tRNA nucleotide sequences: #336 (Thanks @YoungerRen)
  • Replacement of Circos by pyCirclize: #344 (Thanks @acarafat / @alexweisberg)
  • Introduce --label, --size and --dpi parameters in bakta_plot to customize plots: #344
  • Add fixes for common submission errors: #330 (Thanks @kevinmyers)
  • Improve INSDC compliance: 30b3928 8ba8a14
  • Reduce the locus tag offset from 5 to 1 and allow specifications via --locus-tag-increment: #279 (Thanks @1996xjm)
  • Revise internal/external designation of truncated features and (real) pseudogenes: #333
  • Print msg if IPS and pseudogene detection is skipped using the db light version: #320 (Thanks @flashton2003)
  • Review internal main data structure: #338

Fixes

  • Fix import of user-provided CDS spanning sequence edges: #324 #332 (Thanks @aekazakov)
  • Fix translated protein sequences for CDS truncated at both sides: #340 (Thanks @tpillone)
  • Fix wrong truncation label: #341
  • Fix import of user-provided regions in combination with a metadata replicon table: #326

Full Changelog: v1.9.4...v1.10.0