examples/02_process_mri_data.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# MRI data analysis with `cloudknot`\n",
    "\n",
    "This example demonstrates analysis of MRI data using software that depends on the scipy stack, \n",
    "as well as on smaller open-source software projects\n",
    "\n",
    "The data for each participant in the study is non-trivial in size (about 140 MB per subject),\n",
    "and complexity (4D arrays representing MRI measurements in every location in the brain with \n",
    "diffusion gradients in multiple directions). The analysis requires non-trivial computations \n",
    "(e.g., fitting a linear model over directions in every spatial location). \n",
    "\n",
    "The data we will use is publicly accessible through a URL pointing to the Stanford Data Repository\n",
    "We will pull down each participant's data using a URL that the function constructs based on its \n",
    "input."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import cloudknot as ck"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def calculate_fa(sub):\n",
    "    \"\"\" \n",
    "    Process MRI data from a single subject. \n",
    "    \n",
    "    Saves results to S3.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "\n",
    "    sub : str\n",
    "        The ID string forone of the subjects in this data-set. \n",
    "        One of  {'SUB1', 'SUB2', 'SUB3', 'SUB4', 'SUB5', 'SUB6'}\n",
    "    \n",
    "    \"\"\"\n",
    "    import os\n",
    "    import os.path as op\n",
    "    import requests\n",
    "\n",
    "    # Helper function to download files and save them:\n",
    "    def download_file(url, fname):\n",
    "        if not op.exists(fname):\n",
    "            r = requests.get(url)\n",
    "            with open(fname , 'wb') as fd:\n",
    "                for chunk in r.iter_content(chunk_size=128):\n",
    "                    fd.write(chunk)\n",
    "    \n",
    "    print(\"Creating folder structure for data\")\n",
    "    base_folder = op.join(op.expanduser('~'), 'data')\n",
    "    if not op.exists(base_folder):\n",
    "        os.mkdir(base_folder)\n",
    "        os.mkdir(op.join(base_folder, 'sub-01'))\n",
    "        os.mkdir(op.join(base_folder, 'sub-01', 'sess-01'))\n",
    "        anat_folder = op.join(base_folder, 'sub-01', 'sess-01', 'anat')\n",
    "        os.mkdir(anat_folder)\n",
    "        dwi_folder = op.join(base_folder, 'sub-01', 'sess-01', 'dwi')\n",
    "        os.mkdir(dwi_folder)\n",
    "\n",
    "    print(\"Downloading data from the Stanford Data Repository\")\n",
    "    data_url = 'https://stacks.stanford.edu/file/druid:rt034xr8593/'\n",
    "    sub_data = data_url + sub\n",
    "    fbvals_url = sub_data + \"_1.bvals\"\n",
    "    fbvecs_url = sub_data + \"_1.bvecs\"\n",
    "    fnii_url = sub_data + \"_1.nii.gz\"\n",
    "    \n",
    "    download_file(fnii_url, op.join(dwi_folder, 'dwi.nii.gz'))\n",
    "    download_file(fbvals_url, op.join(dwi_folder, 'dwi.bvals'))\n",
    "    download_file(fbvecs_url, op.join(dwi_folder, 'dwi.bvecs'))\n",
    "\n",
    "    print(\"Creating AFQ object\")   \n",
    "    from AFQ.api import AFQ\n",
    "    my_afq = AFQ(preproc_path=base_folder, sub_prefix='sub')\n",
    "    \n",
    "    print(\"Calculating FA\")\n",
    "    fa_file = my_afq.dti_fa[0]\n",
    "\n",
    "    print(\"Uploading to S3\")\n",
    "    client = boto3.resource('s3')\n",
    "    bucket_name = 'escience.washington.edu.public'\n",
    "    b = client.Bucket(bucket_name)\n",
    "    b.upload_file(fa_file, '%s_FA.nii.gz' % sub)    \n",
    "    print(\"Return FA back to user\")\n",
    "    import nibabel as nib\n",
    "    fa = nib.load(fa_file).get_data()\n",
    "    return fa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "knot = ck.Knot(name='calculate-fa',\n",
    "               # With image_github_installs, we can select a fork/branch to install from:\n",
    "               image_github_installs=(\"https://github.com/arokem/pyAFQ.git@no-anat\"),\n",
    "               base_image=\"python:3.7\",\n",
    "               # Or even a specific commit:\n",
    "               #image_github_installs=(\"https://github.com/arokem/pyAFQ.git@afd7221\"),\n",
    "               func=calculate_fa, pars_policies=('AmazonS3FullAccess',))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "result_futures = knot.map(['SUB1', 'SUB2', 'SUB3', 'SUB4', 'SUB5', 'SUB6'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Job ID              Name                        Status   \n",
      "---------------------------------------------------------\n",
      "d43dc89b-745e-4b6b-80ab-e53c4438f688        calculate_fa-5              SUCCEEDED\n",
      "33262a8b-91ec-48f5-ba3a-a95c1a5a21bc        calculate_fa-4              SUCCEEDED\n",
      "49f4e299-bf36-45b6-a7c8-af0446d64f6e        calculate_fa-2              SUCCEEDED\n",
      "0314b338-4b3f-4ff9-9b9b-e211cb51f2ae        calculate_fa-3              SUCCEEDED\n",
      "799d06f5-37fe-460b-802d-ca87b92f3aaf        calculate_fa-0              SUCCEEDED\n",
      "56fd41c1-6841-4f83-ab9e-08ed36251a94        calculate_fa-1              SUCCEEDED\n"
     ]
    }
   ],
   "source": [
    "knot.view_jobs()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "r0 = result_futures[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "fa0 = r0.result()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1120a68d0>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVYAAAECCAYAAAClsP9iAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDcMwhsz/ARAeyV0ZjkjiAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x111ed7630>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.matshow(fa0[:, :, fa0.shape[-1]//2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#knot.clobber(clobber_pars=True, clobber_repo=True, clobber_image=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}