Closed binarybottle closed 7 years ago
I wrote a function that concatenates strings (outputs of nipype's recon-all or antsCorticalThickness.sh functions) to construct paths for the mindboggle123 nipype pipeline, which is now working.
Now I need to figure out how to use ENTRYPOINT in the Dockerfile.mindboggle.complete so that I can 'docker run mindboggle' with arguments properly.
outputs of nodes are not available this way.
subjects_dir = reconall.outputs.subjects_dir
subject_id = reconall.outputs.subject_id
the outputs function simply lists the output names rather than the values.
you would connect the outputs from recon-all and ants segmentation to your mindboggle function. you don't need the concatenate function in the normal nipype connect way. and then you can do whatever you want inside that function to create the mindboggle call.
I removed the extraneous lines from mindboggle123, which doesn't rely on them to connect reconall and corticalthickness nodes to the mindboggle node, but instead relies on an intermediary node that joins path strings:
mbFlow.connect(JoinStrings1, 'new_string', Mindboggle, 'DATA') mbFlow.connect(JoinStrings2, 'new_string', Mindboggle, 'ants')
i don't see the need for JoinStringsX
at all. why couldn't you do that in the mindboggle function?
which can take as input (fsdir, subjectid, antsseg) and do the things joinstrings is doing.
Of course! Thank you -- fixed.
The Docker container can now be run as an executable with:
docker run --rm -ti -v $HOST:/home/jovyan/work nipy/mindboggle $IMAGE --id $ID --out $OUT
with the following in the Dockerfile.mindboggle.complete:
ENTRYPOINT ["mindboggle123"] CMD ["IMAGE", "--id", "--out"]
I have created a draft for a simple nipype pipeline that will -- with the single command "mindboggle123" -- run recon-all, antsCorticalThickness.sh, and mindboggle. However, I don't know how to connect nipype's ReconAll and antsCorticalThickness nodes to a function node I created for mindboggle. See:
https://github.com/nipy/mindboggle/blob/master/mindboggle/mindboggle123#L197-L201