{
"cells": [
{
"attachments": {},
"cell_type": "markdown",
"id": "22b3679d",
"metadata": {
"colab_type": "text",
"id": "view-in-github"
},
"source": [
"\n",
"![\"Open](\"https://colab.research.google.com/assets/colab-badge.svg\")
"
]
},
{
"attachments": {},
"cell_type": "markdown",
"id": "843a313c",
"metadata": {},
"source": [
"## FSL's fMRI Data Analysis via Nipype\n",
"\n",
"Author: Monika Doerig\n",
"\n",
"Citation: \n",
"- Chen Y, Hopp FR, Malik M, Wang PT, Woodman K, Youk S and Weber R (2022) Reproducing FSL's fMRI data analysis via Nipype: Relevance, challenges, and solutions. Front. Neuroimaging 1:953215. doi: [10.3389/fnimg.2022.953215](https://www.frontiersin.org/articles/10.3389/fnimg.2022.953215/full)\n",
"\n",
"- [Original code on OSF](https://osf.io/prg53/?view_only=9d7974834a484cdb972bcc3989589b78)"
]
},
{
"attachments": {},
"cell_type": "markdown",
"id": "a6aa2b67",
"metadata": {},
"source": [
"## Setup Neurodesk"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "db232ff4",
"metadata": {},
"outputs": [],
"source": [
"%%capture\n",
"import os\n",
"import sys\n",
"IN_COLAB = 'google.colab' in sys.modules\n",
"\n",
"if IN_COLAB:\n",
" os.environ[\"LD_PRELOAD\"] = \"\";\n",
" os.environ[\"APPTAINER_BINDPATH\"] = \"/content,/tmp,/cvmfs\"\n",
" os.environ[\"MPLCONFIGDIR\"] = \"/content/matplotlib-mpldir\"\n",
" os.environ[\"LMOD_CMD\"] = \"/usr/share/lmod/lmod/libexec/lmod\"\n",
"\n",
" !curl -J -O https://raw.githubusercontent.com/NeuroDesk/neurocommand/main/googlecolab_setup.sh\n",
" !chmod +x googlecolab_setup.sh\n",
" !./googlecolab_setup.sh\n",
"\n",
" os.environ[\"MODULEPATH\"] = ':'.join(map(str, list(map(lambda x: os.path.join(os.path.abspath('/cvmfs/neurodesk.ardc.edu.au/neurodesk-modules/'), x),os.listdir('/cvmfs/neurodesk.ardc.edu.au/neurodesk-modules/')))))"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "0bc44d6c-c479-4071-bf4b-8e67c9700786",
"metadata": {
"tags": []
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"vendor_id\t: AuthenticAMD\n",
"model name\t: AMD EPYC 7742 64-Core Processor\n"
]
}
],
"source": [
"# Output CPU information:\n",
"!cat /proc/cpuinfo | grep 'vendor' | uniq\n",
"!cat /proc/cpuinfo | grep 'model name' | uniq"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "ca8670a9",
"metadata": {},
"outputs": [],
"source": [
"!pip install pandas nilearn matplotlib"
]
},
{
"attachments": {},
"cell_type": "markdown",
"id": "beb8b3fa",
"metadata": {},
"source": [
"# First-level GLM using Nipype FSL\n",
"\n",
"In this notebook, we recreate the first-level GLM and the first level GLM of FSL GUI using nipype code. For each nipype node, we list the corresponding fsl command from the log file. The dataset we use is a Flanker task, which can be downloaded [here](https://openneuro.org/datasets/ds000102/versions/00001).\n",
"\n",
"We also borrow some helps from this [document](https://nipype.readthedocs.io/en/latest/users/examples/fmri_fsl.html). \n"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "b2bbe67d-00d1-4781-9ad1-fac8f3905356",
"metadata": {
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"['fsl/6.0.5.1']"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#load fsl module\n",
"import lmod\n",
"await lmod.purge(force=True)\n",
"await lmod.load('fsl/6.0.5.1') #Original pipeline: FSL 6.0.4, nipype 1.6.1.\n",
"await lmod.list()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "4abe0d95",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"import os\n",
"os.environ[\"FSLOUTPUTTYPE\"]=\"NIFTI_GZ\""
]
},
{
"attachments": {},
"cell_type": "markdown",
"id": "82d81839",
"metadata": {},
"source": [
"## Preparation\n",
"Import all the relevant libraries needed for the preprocessing stage."
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "bf70c656",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"from __future__ import print_function\n",
"from __future__ import division\n",
"from builtins import str\n",
"from builtins import range\n",
"\n",
"from glob import glob\n",
"\n",
"from nipype import Function\n",
"from nipype.interfaces import fsl, utility as util, io as nio\n",
"import nipype.pipeline.engine as pe # pypeline engine\n",
"import nipype.algorithms.modelgen as model # model generation\n",
"import nipype.algorithms.rapidart as ra # artifact detection\n",
"import matplotlib.pyplot as plt\n",
"\n",
"import nilearn\n",
"from nilearn import surface\n",
"from nilearn import plotting"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "6d9cb873",
"metadata": {
"tags": []
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"NIFTI_GZ\n",
"6.0.5.1:57b01774\n"
]
}
],
"source": [
"# check output type and fsl version\n",
"print(fsl.Info.output_type())\n",
"print(fsl.Info.version())"
]
},
{
"attachments": {},
"cell_type": "markdown",
"id": "fad038b8",
"metadata": {},
"source": [
"#### Set up data path"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "a8005f06",
"metadata": {
"scrolled": true,
"tags": []
},
"outputs": [],
"source": [
"# Input: Set the BIDS path\n",
"data_dir_first = os.path.join(os.getcwd(), 'ds000102')\n",
"# Output: Set path where nipype will store stepwise results\n",
"exp_dir_first = os.path.join(os.getcwd(), 'output_level1')\n",
"\n",
"# Input: Set path of first level outputs\n",
"data_dir_second = os.path.join(os.getcwd(), 'output_level1/level1_results/')\n",
"# Output: Set path where nipype will store stepwise results\n",
"exp_dir_second = os.path.join(os.getcwd(), 'output_level2/')\n",
"\n",
"# Input: Set path of second level outputs\n",
"data_dir_third = os.path.join(os.getcwd(), 'output_level2/level2_results/')\n",
"# Output: Set path where nipype will store stepwise results\n",
"exp_dir_third = os.path.join(os.getcwd(), 'output_level3/')\n",
"\n",
"PATTERN = \"sub-0*\""
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "13564f7b",
"metadata": {
"editable": true,
"scrolled": true,
"slideshow": {
"slide_type": ""
},
"tags": [
"scroll-output"
]
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Clone attempt: 0%| | 0.00/2.00 [00:00 [ [ ]]