support detection of RBS across linear chromosome breakpoint ?

Hi @althonos ,

I hope you've been well mate! I have an issue/discussion to raise based on some related discussions here on my circular contig reorientation tool Dnaapler https://github.com/gbouras13/dnaapler/issues/90 . The default and desired behaviour of dnaapler is to reorient the chromosome to begin with the start codon of dnaA at coordinate 1 (there is some unrelated discussion in that issue about issues with reorientation that you can ignore :) ).

However, @oschwengers makes the very good point that for bacterial chromosomes that are re-oriented this way, when pyrodigal is run (e.g. in annotation with bakta), it may fail to call the dnaA gene properly, as the RBS will be on the other end of the (linearised) circular contig.

I am personally not a fan of changing dnaapler so that it offsets starting with dnaA with e.g. 50 or 100bps like Oliver is suggesting (due to the diversity in RBSs and intergenic distances before dnaA across all bacteria), and therefore I propose that the best answer would be for prodigal to detect RBSs across edges (if you haven't already implemented such functionality) - is this possible in you view (and of course would you want to implement it)? And if not, do you have other suggestions?

George