Permalink
Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
57 lines (48 sloc) 3.44 KB

CEGMA RELEASE NOTES

v2.5 - 2014-05-19

  • Fixed qw() bug in foreach loops of two required Perl modules. Perl 5.13 and later no longer allows you use qw() to form lists in for/foreach loops without enclosing function in another set of parentheses.
  • Fixed very rare bug that could have ignored a possible core gene if only 1 of the 6 core proteins for any KOG matched a genomic region. This occurs in the TBLASTN step which looks to produce a list of 5 potential candidate regions for each KOG. Regions that are only supported by a KOG protein from a single species are highly unlikely to correspond to real core genes, but theoretically it was possible that this bug could cause 1 or 2 core genes to be missed.
  • Changed use of -max_num_description option for tblastn to instead use -max_target_seqs.
  • Fixed typo in documentation which mentioned -hmm_directory option instead of -hmm_profiles.
  • Use of -v|--verbose option now gives much more useful progress information about how far CEGMA has progressed through each stage.
  • More detailed output is produced for some steps when using -v|--verbose mode.
  • Have moved some functions that are used by multiple scripts to a new Cegma.pm module.
  • Tried improving (and harmonizing) various error and warning messages.
  • Added more graceful ways of ending the script prematurely if no core genes are found at any stage of analysis (rather than just dying with a cryptic error message).
  • Removed duplication of exon details in GFF output. Lines with GFF feature description of 'Exon' are no longer printed (just 'First', 'Internal', and 'Terminal').
  • Added more default information to completeness.pl output, and refined use of -v|--verbose option.
  • Updated README file to include a new section about citing CEGMA and added link to the new CEGMA FAQ
  • Moved code to a GitHub repository
  • Included release notes file as part of code distribution

v2.4

  • Fixed use of -threads option
  • Fixed bug which happened when genome contained no CEGs at all
  • Fixed error that could occur when processing some unusual genewise predictions
  • Fixed bug in calculating completeness statistics for the subset of 248 highly conserved CEGs
  • Made some error messages a little more meaningful
  • Made hmmsearch produce more useful verbose output

v2.3.1

  • Added --threads details to built-in documentation
  • Used 'config no_ignore_case' options for Getopt::Long to avoid case sensitivity problems with -t/-T options
  • Fixed bug where if no protein alignments were made by genewise, we would not run geneid but still try to run awk on a nonexistent geneid output file.
  • Made some error messages a little more useful when script dies
  • parsewise script contained a bug that would manifest if a genewise prediction did not start with a loop state. The code to fix this scenario would not be applied if the last genewise prediction did not start with a loop.
  • Fixed ortholog statistics generated by final step of pipeline
  • Added key to final table

v2.3

  • Supports NCBI BLAST+ only (support for WU-BLAST has been removed)
  • New --threads option to let you utliize multiple cores when running tblastn and hmmsearch
  • Now works with latest versions of HMMER and geneid
  • Misc. small improvements to code