Skip to content

Latest commit

 

History

History
54 lines (44 loc) · 1.98 KB

CHANGELOG.rst

File metadata and controls

54 lines (44 loc) · 1.98 KB

Hey, let's keep a change log.

next

0.10.0

  • Added things.Region class for generic genomic regions
  • Updated dependency information for latest Ubuntu
  • Added ORF class and mRNA.longest_orf() method
  • Added script to filter FASTA files from abundance estimation (Salmon) results
  • Added new script for classifying gene coverage in new assemblies
  • Exports to GFF3 updated: tRNAs include anticodon, rRNAs includ product name
  • mRNA.longest_orf() now has require_start option
  • fasta_simple_stats.py now exports N50 and N90 stats
  • fasta_simple_stats.py no longer errors on blank lines

0.9.0

  • Added ncRNA feature class and changed inheritance for all related types (rRNA, tRNA, etc.)
  • gff.parse_annotation_from_column_9() now recognizes both 'product_name' and 'product' in 9th column.

0.8.0

  • This actually should have been what the previous release was called. X.X.N releases should only be for bug fixes.
  • Added dependency support for R in Docker build
  • Added report_go_slim_counts.py script
  • Added metaphlan-parsing R script (though needs major clean-up)

0.7.1

  • NCBI API key now required for all E-Utility calls
  • Added --skip_existing option to batch genbank downloader" download_assemblies_from_genbank.py
  • Changed genbank/download_assemblies_from_genbank.py to export BOTH contig and scaffold range files automatically
  • Bugfix: genbank module was failing if the gene symbol contained a comma
  • Corrected python module dependency from igraph -> python-igraph
  • Updated documentation on taxadb generation in general/filter_uniref_by_taxonomy.py
  • Added taxadb to formal dependencies
  • Changed BED export library to use gene.id for 4th column if no locus is found
  • Added ncbigff module and script to use it for conversions
  • Added general/join_columnar_files.py
  • Added fasta/fasta_simple_stats.py

0.7.0

  • Added script to convert GFF3 to BED
  • Added script to convert BLAST/RAPSearch2 to BED
  • Added script to convert HMMer3 HTAB to BED
  • Added biocode.bed library