Docs from RST to Markdown

README.md

@@ -0,0 +1,46 @@
+[![Documentation Status](https://readthedocs.org/projects/hippunfold/badge/?version=latest)](https://hippunfold.readthedocs.io/en/latest/?badge=latest)
+[![CircleCI](https://circleci.com/gh/khanlab/hippunfold.svg?style=svg)](https://circleci.com/gh/khanlab/hippunfold)
+![PyPI](https://img.shields.io/pypi/v/hippunfold)
+![Docker Pulls](https://img.shields.io/docker/pulls/khanlab/hippunfold)
+
+# Hippunfold
+
+This tool aims to automatically model the topological folding structure
+of the human hippocampus, and computationally unfold the hippocampus to 
+segment subfields and generate hippocampal and dentate gyrus surfaces.
+
+The overall workflow can be summarized in the following steps:
+
+![Pipeline Overview](https://raw.githubusercontent.com/khanlab/hippunfold/master/docs/images/hippunfold_overview.jpg)
+
+1.  Pre-processing MRI images including non-uniformity correction, 
+    resampling to 0.3mm isotropic subvolumes, registration and cropping to coronal-oblique 
+    subvolumes around each hippocampus
+2.  Automatic segmentation of hippocampal tissues and surrounding
+    structures via deep convolutional neural network U-net (nnU-net), models available
+    for T1w, T2w, hippocampal b500 dwi, and neonatal T1w, and post-processing
+    with fluid label-label registration to a high-resolution average template
+3.  Imposing of coordinates across the anterior-posterior and 
+    proximal-distal dimensions of hippocampal grey matter
+    via solving Laplace's equation, and using an equivolume solution for
+    laminar coordinates
+4.  Generating transformations between native and unfolded spaces using scattered 
+    interpolation on the hippocampus and dentate gyrus coordinates separately
+5.  Applying these transformations to generate surfaces meshes of the hippocampus
+    and dentate gyrus, and extraction of morphological surface-based features including
+    thickness, curvature, and gyrification index, sampled at the midthickness surface, and mapping 
+    subfield labels from the histological BigBrain atlas of the hippocampus 
+6.  Generating high-resolution volumetric segmentations of the subfields using the same
+    transformations and volumetric representations of the coordinates.
+
+
+**Full Documentation:**  [here](https://hippunfold.readthedocs.io**)
+
+
+**Relevant papers:**
+-  DeKraker J, Haast RAM, Yousif MD, Karat B, Köhler S, Khan AR. HippUnfold: Automated hippocampal unfolding, morphometry, and subfield segmentation. bioRxiv 2021.12.03.471134; doi: 10.1101/2021.12.03.471134 [link](https://www.biorxiv.org/content/10.1101/2021.12.03.471134v1)
+- DeKraker J, Ferko KM, Lau JC, Köhler S, Khan AR. Unfolding the hippocampus: An intrinsic coordinate system for subfield segmentations and quantitative mapping. Neuroimage. 2018 Feb 15;167:408-418. doi: 10.1016/j.neuroimage.2017.11.054. Epub 2017 Nov 23. PMID: 29175494. [link](https://pubmed.ncbi.nlm.nih.gov/29175494/)
+- DeKraker J, Lau JC, Ferko KM, Khan AR, Köhler S. Hippocampal subfields revealed through unfolding and unsupervised clustering of laminar and morphological features in 3D BigBrain. Neuroimage. 2020 Feb 1;206:116328. doi: 10.1016/j.neuroimage.2019.116328. Epub 2019 Nov 1. PMID: 31682982. [link](https://pubmed.ncbi.nlm.nih.gov/31682982/)
+- DeKraker J, Köhler S, Khan AR. Surface-based hippocampal subfield segmentation. Trends Neurosci. 2021 Nov;44(11):856-863. doi: 10.1016/j.tins.2021.06.005. Epub 2021 Jul 22. PMID: 34304910. [link](https://pubmed.ncbi.nlm.nih.gov/34304910/)
+
+

docs/conf.py

@@ -32,8 +32,12 @@
 extensions = [
     "sphinx_rtd_theme",
     "sphinxarg.ext",
+    'm2r2',
 ]
 
+source_suffix = ['.rst', '.md']
+m2r_parse_relative_links = True
+
 # Add any paths that contain templates here, relative to this directory.
 templates_path = ['_templates']
 

docs/contributing/contributing.md

@@ -0,0 +1,68 @@
+# Contributing to Hippunfold
+
+Hippunfold dependencies are managed with Poetry, which you\'ll need
+installed on your machine. You can find instructions on the [poetry
+website](https://python-poetry.org/docs/master/#installation).
+
+## Set-up your development environment:
+
+Clone the repository and install dependencies and dev dependencies with
+poetry:
+
+    git clone http://github.com/khanlab/hippunfold
+    cd hippunfold
+    poetry install
+
+Poetry will automatically create a virtualenv. To customize \... (TODO:
+finish this part)
+
+Then, you can run hippunfold with:
+
+    poetry run hippunfold
+
+or you can activate a virtualenv shell and then run hippunfold directly:
+
+    poetry shell
+    hippunfold
+
+You can exit the poetry shell with `exit`.
+
+## Running code format quality checking and fixing:
+
+Hippunfold uses [poethepoet](https://github.com/nat-n/poethepoet) as a
+task runner. You can see what commands are available by running:
+
+    poetry run poe
+
+We use `black` and `snakefmt` to ensure
+formatting and style of python and Snakefiles is consistent. There are
+two task runners you can use to check and fix your code, and can be
+invoked with:
+
+    poetry run poe quality_check
+    poetry run poe quality_fix
+
+Note that if you are in a poetry shell, you do not need to prepend
+`poetry run` to the command.
+
+## Dry-run testing your workflow:
+
+Using Snakemake\'s dry-run option (`--dry-run`/`-n`) is an easy way to verify any
+changes to the workflow are working correctly. The `test_data` folder contains a 
+number of *fake* bids datasets (i.e. datasets with zero-sized files) that are useful
+for verifying different aspects of the workflow. These dry-run tests are
+part of the automated github actions that run for every commit.
+
+You can use the hippunfold CLI to perform a dry-run of the workflow,
+e.g. here printing out every command as well:
+
+    hippunfold test_data/bids_singleT2w test_out participant --modality T2w -np
+
+As a shortcut, you can also use `snakemake` instead of the
+hippunfold CLI, as the `snakebids.yml` config file is set-up
+by default to use this same test dataset, as long as you run snakemake
+from the `hippunfold` folder that contains the
+`workflow` folder:
+
+    cd hippunfold
+    snakemake -np

docs/getting_started/installation.md

@@ -0,0 +1,167 @@
+# Installation
+
+BIDS App for Hippocampal AutoTop (automated hippocampal unfolding and
+subfield segmentation)
+
+## Requirements
+
+-   Docker (Mac/Windows/Linux) or Singularity (Linux)
+-   `pip install` is also available (Linux), but will still require
+    singularity to handle some dependencies
+-   For the T1w workflow: BIDS dataset with T1w images.
+    Higher-resolution data are preferred (\<= 0.8mm) but the pipeline
+    will still work with 1mm T1w images.
+-   GPU not required
+
+### Notes:
+
+-   T1w and/or T2w input images are supported, and we recommend using
+    sub-millimetric isotropic data for best performance.
+-   Other 3D imaging modalities (eg. ex-vivo MRI, 3D histology, etc.)
+    can be used, but may require manual tissue segmentation.
+
+## Running with Docker
+
+Pull the container:
+
+    docker pull khanlab/hippunfold:latest
+
+Do a dry run, printing the command at each step:
+
+    docker run -it --rm \
+    -v PATH_TO_BIDS_DIR:/bids:ro \
+    -v PATH_TO_OUTPUT_DIR:/output \
+    khanlab/hippunfold:latest \
+    /bids /output participant -np 
+
+Run it with maximum number of cores:
+
+    docker run -it --rm \
+    -v PATH_TO_BIDS_DIR:/bids:ro \
+    -v PATH_TO_OUTPUT_DIR:/output \
+    khanlab/hippunfold:latest \
+    /bids /output participant -p --cores all
+
+For those not familiar with Docker, the first three lines of this
+example are generic Docker arguments to ensure it is run with the safest
+options and has permission to access your input and output directories
+(specified here in capital letters). The third line specifies the
+HippUnfold Docker container, and the fourth line contains the requires
+arguments for HippUnfold. You may want to familiarize yourself with
+[Docker options](https://docs.docker.com/engine/reference/run/), and an
+overview of HippUnfold arguments is provided in the [Command line
+interface](https://hippunfold.readthedocs.io/en/latest/usage/app_cli.html)
+documentation section.
+
+## Running with Singularity
+
+If you are using singularity, there are two different ways to run
+`hippunfold`.
+
+1.  Pull the fully-contained BIDS app container from
+    `docker://khanlab/hippunfold:latest` and run it
+2.  Clone the github repo, `pip install`, and run `hippunfold`, which
+    will pull the required singularity containers as needed.
+
+If you want to modify the workflow at all or run a development branch,
+then option 2 is required.
+
+Furthermore, if you want to run the workflow on a Cloud system that
+Snakemake supports, or with a Cluster Execution profile, you should use
+Option 2 (see snakemake documentation on [cluster
+profiles](https://github.com/snakemake-profiles/doc) and [cloud
+execution](https://snakemake.readthedocs.io/en/stable/executing/cloud.html)).
+
+### Option 1:
+
+Pull the container:
+
+    singularity pull khanlab_hippunfold_latest.sif docker://khanlab/hippunfold:latest
+
+Do a dry run, printing the command at each step:
+
+    singularity run -e khanlab_hippunfold_latest.sif \
+    PATH_TO_BIDS_DIR PATH_TO_OUTPUT_DIR participant -np 
+
+Run it with maximum number of cores:
+
+    singularity run -e khanlab_hippunfold_latest.sif \
+    PATH_TO_BIDS_DIR PATH_TO_OUTPUT_DIR participant -p --cores all
+
+### Option 2:
+
+This assumes you have created and activated a virtualenv or conda
+environment first.
+
+1.  Install hippunfold using pip:
+
+        pip install hippunfold
+
+2.  Run the following to download the U-net models:
+
+        hippunfold_download_models
+
+3.  Now hippunfold will be installed for you and can perform a dry-run
+    with:
+
+        hippunfold  PATH_TO_BIDS_DIR PATH_TO_OUTPUT_DIR participant -np
+
+4.  To run on all cores, and have snakemake pull any required
+    containers, use:
+
+        hippunfold  PATH_TO_BIDS_DIR PATH_TO_OUTPUT_DIR participant --cores all --use-singularity
+
+## Additional instructions for Compute Canada
+
+This section provides an example of how to set up a `pip installed` copy
+of HippUnfold on CompateCanada\'s `graham` cluster.
+
+### Setting up a dev environment on graham:
+
+Here are some instructions to get your python environment set-up on
+graham to run HippUnfold:
+
+1.  Create a virtualenv and activate it:
+
+        mkdir $SCRATCH/hippdev
+        cd $SCRATCH/hippdev
+        module load python/3
+        virtualenv venv
+        source venv/bin/activate
+
+2.  Follow the steps above to `pip install` from github repository
+
+### Running hippunfold jobs on graham:
+
+Note that this requires
+[neuroglia-helpers](https://github.com/khanlab/neuroglia-helpers) for
+regularSubmit or regularInteractive wrappers, and the
+[cc-slurm](https://github.com/khanlab/cc-slurm) snakemake profile for
+graham cluster execution with slurm.
+
+In an interactive job (for testing):
+
+    regularInteractive -n 8
+    hippunfold PATH_TO_BIDS_DIR PATH_TO_OUTPUT_DIR participant \
+    --participant_label 001 -j 8
+
+Here, the last line is used to specify only one subject from a BIDS
+directory presumeably containing many subjects.
+
+Submitting a job (for larger cores, more subjects), still single job,
+but snakemake will parallelize over the 32 cores:
+
+    regularSubmit -j Fat \
+    hippunfold PATH_TO_BIDS_DIR PATH_TO_OUTPUT_DIR participant  -j 32
+
+Scaling up to \~hundred subjects (needs cc-slurm snakemake profile
+installed), submits 1 16core job per subject:
+
+    hippunfold PATH_TO_BIDS_DIR PATH_TO_OUTPUT_DIR participant \
+    --profile cc-slurm
+
+Scaling up to even more subjects (uses group-components to bundle
+multiple subjects in each job), 1 32core job for N subjects (e.g. 10):
+
+    hippunfold PATH_TO_BIDS_DIR PATH_TO_OUTPUT_DIR participant \
+    --profile cc-slurm --group-components subj=10