...
Hail is available as part of the hr32 software project within the NCI-bio-python module.
On
...
On OOD there is no need to start a separate Spark cluster for hail to use. Instead hail will start a local Spark cluster that uses all the cores that were provisioned for the Jupyterlab session. The NCI-bio-python module needs to be loaded when starting Jupyterlab (see Bioinformatics Analysis (hr32) for more details).
To start hail:
| Code Block | ||
|---|---|---|
| ||
import hail as hl
# Initialise hail and start a local spark cluster
hl.init()
# For a list of arguments that can be passed to hail, type: help(hl.init) |
On gadi
On gadi there are two options for running hail:
...