linsalrob / PhiSpy

Prediction of prophages from bacterial genomes
MIT License
70 stars 20 forks source link

"Contig starting with a partial ORF" warning #60

Open EGTortuero opened 1 year ago

EGTortuero commented 1 year ago

Dear Rob,

I am currently using PhiSpy version 4.2.21 to detect prophages in several strains of Aeromonas and, for one of them (that is available at the NCBI: https://www.ncbi.nlm.nih.gov/nuccore/CP043323.1/) and it prints this warning concerning an untested feature:

$ phispy --output_choice 512 -o CP043323_prophages CP043323.1.gbk
Feature FT688_00005 spans the origin: join{[0:44](-), [4784838:4786216](-)}
WARNING: THIS IS AN UNTESTED FEATURE!
We have not thoroughly tested conditions where an ORF on the -ve strand appears to cross the origin of the contig. We would appreciate you posting an issue on GitHub and sending Rob a copy of your genome to test!
Feature FT688_00005 spans the origin: join{[0:44](-), [4784838:4786216](-)}
WARNING: THIS IS AN UNTESTED FEATURE!
We have not thoroughly tested conditions where an ORF on the -ve strand appears to cross the origin of the contig. We would appreciate you posting an issue on GitHub and sending Rob a copy of your genome to test!
Processing 1 contigs
Making Testing Set...

Since the fact that some available genomes might have the first coordinate in the "middle" of an ORF, I was wondering if there is any chance of missing a potential prophage in these situations. If so, what would you recommend at this moment to check that there is nothing lost? Finally, and the most relevant question, would you have planned to improve PhiSpy to deal with situations like this one, please?

Best,

Enrique