I noticed that sometimes when running the pipeline I'd get this warning from BLAST: [blastp] Query is Empty!
The issue ended up being that the pipeline wasn't handling cases where the contig doesn't have any ORFs (this basically only happens when a contig is really small). It would try to run the first step of the pipeline anyway, resulting in this warning.
This fixes the issue, and also refactors some of pipeline.run and adds a couple more methods to basically reset instance variables before trying to process new data. I originally thought that the pipeline was erroneously trying to reuse stale data (it wasn't) but these methods probably should have existed anyway and make pipeline.run a bit cleaner.
I noticed that sometimes when running the pipeline I'd get this warning from BLAST:
[blastp] Query is Empty!The issue ended up being that the pipeline wasn't handling cases where the contig doesn't have any ORFs (this basically only happens when a contig is really small). It would try to run the first step of the pipeline anyway, resulting in this warning.
This fixes the issue, and also refactors some of
pipeline.runand adds a couple more methods to basically reset instance variables before trying to process new data. I originally thought that the pipeline was erroneously trying to reuse stale data (it wasn't) but these methods probably should have existed anyway and makepipeline.runa bit cleaner.