This docker wraps OBITools (1.2.13) to make the workflows a bit more portable.
The following approach calls obitools functions from Python:
import docker
from docker.errors import NotFound
client = docker.from_env()
# Create or start docker container named 'obitools'.
try:
obt = client.containers.get('obitools')
if obt.status == 'exited':
print('Starting container obitools.')
obt.start()
else:
print('Container obitools ready.')
except NotFound:
print('Creating container obitools.')
obt = client.containers.run(
image='romunov/obitools:1.2.13',
name='obitools',
# This maps my local wolfdata to /data in container
volumes={'/home/romunov/wolfdata': {'bind': '/data', 'mode': 'rw'}},
tty=True, detach=True
)
To run the first command from the wolves tutorial, you would run something along the lines of
reads_r = "wolf_R.fastq"
reads_f = "wolf_F.fastq"
output = "wolf.fastq"
wolf = obt.exec_run(tty=True, workdir='/data', cmd=f'illuminapairedend --score-min={score_min} -r {reads_r} {reads_f} > {output}')
The result would be found in /home/romunov/wolfdata. See the vignette Python notebook for the full workflow.