The current testing uses a subsetted reference genome, (2L, 2-3Mb) which means that coordinates are all off. This causes things like the Kdr origins notebook to fail CI, as it tries to subset based on chrom + physical position.
It would be simpler if we could have normal coordinates in future.
We should download the full reference from vectorbase with wget/curl in the stagein of the CI
The current testing uses a subsetted reference genome, (2L, 2-3Mb) which means that coordinates are all off. This causes things like the Kdr origins notebook to fail CI, as it tries to subset based on chrom + physical position.
It would be simpler if we could have normal coordinates in future.