ndmg

NeuroData’s MR Graphs package, ndmg (pronounced “nutmeg”), is the successor of the MRCAP, MIGRAINE, and m2g pipelines. ndmg combines dMRI and sMRI data from a single subject to estimate a high-level connectome reliably and scalably.

Documentation

Please read the official ndmg docs.

Error Reporting

Experiencing problems? Please open an issue and explain what's happening so we can help.

Acknowledgement

When using this pipeline, please acknowledge us with the citations in the attached bibtex file.

Instructions

The bids/ndmg Docker container enables users to run end-to-end connectome estimation on structural MRI right from container launch. The pipeline requires that data be organized in accordance with the BIDS spec. If the data you wish to process is available on S3 you simply need to provide your s3 credentials at build time and the pipeline will auto-retrieve your data for processing.

To get your container ready to run just follow these steps:

(A) I do not wish to use S3:

• In your terminal, type:

  $ docker pull bids/ndmg

(B) I wish to use S3:

• Add your secret key/access id to a file called credentials.csv in this directory on your local machine. A dummy file has been provided to make the format we expect clear. (This is how AWS provides credentials)
• In your terminal, navigate to this directory and type:

  $ docker build -t <yourhandle>/ndmg .

Now we're ready to launch our instances and process some data!

Like a normal docker container, you can startup your container with a single line. Let's assume I am running this and I wish to use S3, so my container is called gkiar/ndmg. If you don't want to use S3, you can replace gkiar with bids and ignore the S3 related flags for the rest of the tutorial.

I can start my container with:

$ docker run -ti bids/ndmg
usage: ndmg_bids [-h]
                 [--participant_label PARTICIPANT_LABEL [PARTICIPANT_LABEL ...]]
                 [--bucket BUCKET] [--remote_path REMOTE_PATH]
                 bids_dir output_dir {participant}
ndmg_bids: error: too few arguments

We should've noticed that I got an error back suggesting that I didn't properly provide information to our container. Let's try again, with the help flag:

$ docker run -ti bids/ndmg:v4 -h

usage: ndmg_bids [-h]
                 [--participant_label PARTICIPANT_LABEL [PARTICIPANT_LABEL ...]]
                 [--bucket BUCKET] [--remote_path REMOTE_PATH]
                 bids_dir output_dir {participant}

This is an end-to-end connectome estimation pipeline from sMRI and DTI images

positional arguments:
  bids_dir              The directory with the input dataset formatted
                        according to the BIDS standard.
  output_dir            The directory where the output files should be stored.
                        If you are running group level analysis this folder
                        should be prepopulated with the results of the
                        participant level analysis.
  {participant}         Level of the analysis that will be performed. Multiple
                        participant level analyses can be run independently
                        (in parallel) using the same output_dir.

optional arguments:
  -h, --help            show this help message and exit
  --participant_label PARTICIPANT_LABEL [PARTICIPANT_LABEL ...]
                        The label(s) of the participant(s) that should be
                        analyzed. The label corresponds to
                        sub-<participant_label> from the BIDS spec (so it does
                        not include "sub-"). If this parameter is not provided
                        all subjects should be analyzed. Multiple participants
                        can be specified with a space separated list.
  --bucket BUCKET       The name of an S3 bucket which holds BIDS organized
                        data. You must have built your bucket with credentials
                        to the S3 bucket you wish to access.
  --remote_path REMOTE_PATH
                        The path to the data on your S3 bucket. The data will
                        be downloaded to the provided bids_dir on your
                        machine.

Cool! That taught us some stuff. So now for the last unintuitive piece of instruction and then just echoing back commands I'm sure you could've figured out from here: in order to share data between our container and the rest of our machine, we need to mount a volume. Docker does this with the -v flag. Docker expects its input formatted as: -v path/to/local/data:/path/in/container. We'll do this when we launch our container, as well as give it a helpful name so we can locate it later on.

Finally:

docker run -ti --name ndmg_test --rm -v ./data:${HOME}/data bids/ndmg ${HOME}/data/ ${HOME}/data/outputs participant --participant_label 01 -b mybucket -r path/on/bucket/