1.1.1

What's Changed

• Fix html and recursion by @hoelzer in #68

This update adds stability for singularity usage, especially with newer nextflow versions where $baseDir might not work. It also hotfixes an issue with the python recursion depth when combining Roary results and ILPs. The user can now increase the recursion depth; it is not an optimal solution but a hotfix. Also, minor problems with wrong paths in the HTML output were solved. Further, some problems with the new "skip prokka step" feature were solved: when the feature is used their is no renaming of FASTA files happening and thus the _RENAMED suffix does not exist and subsequently no need to remove it in later steps such as the MSA generation.

Closed issues:

• #66
• #65
• #67

Full Changelog: 1.1.0...1.1.1

1.1.0

What's Changed

• Feature skip Prokka by @fischer-hub in #64
• add a function to skip the Prokka annotation and directly provide GFF/FAA inputs from previous Prokka runs additionally to the FASTA input
• this allows the user to skip Prokka and use already available annotation files
• the feature is still wip and needs further testing in case the user provides GFF/FAA files with non-matching columns/content

New Contributors

• @fischer-hub made their first contribution in #64

Full Changelog: 1.0.2...1.1.0

1.0.3

Release to get a Zenodo DOI

1.0.2

What's Changed

• Hot fix for issue #54, skipping tree calc if no core gene MSA by @hoelzer in #55

Full Changelog: 1.0.1...1.0.2

For diverse input genomes, no core gene may exist for each input species. By default, RIBAP only considers core genes found in all input species for the (optional) --tree calculation. With this update, a new parameter --core_perc is introduced that can be used to lower the threshold (default: 1.0 == 100%). See example below and the RIBAP --help:

    --core_perc         Define how many species are required so that a gene is considered a core gene for tree calculation.
                        Per default, RIBAP will only consider genes that were found in all input genomes (100%).
                        However, this can cause tree calculation to stop when there are no such core genes. 
                        You can lower the threshold to include more homologous genes into the tree caclulation. 
                        The total input genome number will be multiplied by this value and rounded down when the 
                        results is <= x.5 (e.g., 28*0.9=25.2 --> 25) and up otherwise (e.g., 28*0.95=26.6 --> 27).
                        All RIBAP groups that are composed of genes from different species equal or greater this number 
                        will be considered in the core gene MSA and tree [default: $params.core_perc]

1.0.1

What's Changed

• Example data by @hoelzer in #50
• Update README.md by @lisa-mariebarf in #51
• remove unnecessary gunzip by @hoelzer in #53

New Contributors

• @lisa-mariebarf made their first contribution in #51

Full Changelog: 1.0.0...1.0.1

Release 1.0.0

What's Changed

First full-release of RIBAP.

Since the last pre-release:

• We have rewritten our ILP modules to be less harddrive-demanding.
  • per default: ILPs and their solutions are not stored anymore. For large datasets, this was waaaay too much.
  • we introduced the flag --keepILPs which can be set in the nextflow run command. Using this won't delete ILPs.
• We exchanged the module to build the phylogenetic tree of the core genome. Now IQ-Tree2 is used instead of RAxML.
  • if a core gene family has 100% sequence similarity between all possible comparisons, it'll be ignored for the tree calculation
• RIBAP now supports mamba additionally to conda with -profile mamba (note: you have to install mamba to make this work)
• Some minor changes are included, like correct naming of output directories and allocating resources on HPC systems.

Full Changelog: 0.7.2...1.0.0

0.7.2

What's Changed

• add auto option for RAXML-NG max CPUs on HPC by @hoelzer in #37
• output whole core gene set tree

Full Changelog: 0.7.1...0.7.2

0.7.1

Fix bug introduced while solving #34 when --fasta input was not working anymore when no --reference is used. Now, RIBAP can run again w/ only FASTA input and no additional GenBank annotation file to guide prokka.

0.7.0

• fix #34
  • add --reference parameter to provide a GenBank-styled annotation file to integrate in prokkas functional annotation via the --proteins param
  • to allow mixed annotation files and also regular prokka runs w/o additional annotation, the --list parameter an be used
  • when directly a GenBank file is provided via --reference foo.gbk, ALL annotations of the input genome FASTAs will be guided by that!
• bump version raxml-ng
• update README

0.6.2

Update versions for GLPK and parallel in Conda and Docker profile.