search

nextgenusfs / funannotate

Eukaryotic Genome Annotation Pipeline
http://funannotate.readthedocs.io
BSD 2-Clause "Simplified" License
320 stars 85 forks source link

funannotate-predict.py: bedtools intersect fails -- negative starts #19

Closed h0ha closed 8 years ago

h0ha commented 8 years ago

Hello! https://github.com/nextgenusfs/funannotate/blob/master/bin/funannotate-predict.py#L658 bedtools says starts less then 1

nextgenusfs commented 8 years ago

Thanks, will fix shortly. I'm at a conference but can push these changes later today - I won't do a full release until I know it is stable, but should be able to just clone the repository.

And I appreciate the help in finding bugs!

h0ha commented 8 years ago

welcome %)

h0ha commented 8 years ago

to be serious, You've done a great work. btw what is the way to reference funannotate?

nextgenusfs commented 8 years ago

Thanks. Right now you can cite the github repo I guess: something like

Palmer JM. 2016. Funannotate: a fungal genome annotation and comparative genomics pipeline. https://github.com/nextgenusfs/funannotate.

I pushed some updates to predict to fit this negative location in bedtools, if you get an error on the other end (which you might) it might be a bit longer of a fix, but doable.

h0ha commented 8 years ago

Thank You. I supposed when intersecting bedtools didn't know anything about contigs sizes. Though it's possible to get errors when extracting sequences :(

nextgenusfs commented 8 years ago

Yes, I just need to load the contig names and sizes into a dictionary and make sure that the coordinates don't go outside those parameters. that should be a more stable fix

nextgenusfs commented 8 years ago

Okay, just pushed a fix. Not sure of stability yet, running the test now, but most likely is okay.

h0ha commented 8 years ago

thank you