Micapipe

https://micapipe.readthedocs.io/en/latest
Admin
Micapipe is a multimodal MRI processing pipeline providing a robust framework to analyze...
Tokens:76,222
Snippets:768
Trust Score:9.6
Update:5 months ago
Show doc for...
Tokens:
Raw
### Verify micapipe installation

Source: https://micapipe.readthedocs.io/en/latest/pages/01.install/index.html

Run the micapipe command with the -help flag to confirm that the installation and environment setup are correct.

```bash
$ micapipe -help

```

--------------------------------

### Example `micapipe_anonymize` Run with All Methods

Source: https://micapipe.readthedocs.io/en/latest/pages/05.micapipe_anonymize/index.html

This example shows how to run `micapipe_anonymize` to apply all anonymization methods (-all) to a specific subject and session. Ensure the output directory is empty before running.

```bash
1micapipe_anonymize -sub 01 -ses 01 -out dataset_anonymized -bids dataset
2-all
```

--------------------------------

### Example Final BIDS Structure with Files

Source: https://micapipe.readthedocs.io/en/latest/pages/05.mic2bids/index.html

Demonstrates the complete BIDS structure after conversion, including NIfTI files and their corresponding JSON metadata for the 'anat' directory.

```text
sub-HC001
└── ses-01
 	├── anat
 	│	├── sub-01_ses-01_acq-inv1_T1map.json
 	│	├── sub-01_ses-01_acq-inv1_T1map.nii.gz
 	│	├── sub-01_ses-01_acq-inv2_T1map.json
 	│	├── sub-01_ses-01_acq-inv2_T1map.nii.gz
 	│	├── sub-01_ses-01_acq-mp2rage_T1map.json
 	│	├── sub-01_ses-01_acq-mp2rage_T1map.nii.gz
 	│	├── sub-01_ses-01_acq-uni_T1map.json
 	│	├── sub-01_ses-01_acq-uni_T1map.nii.gz
 	│	├── sub-01_ses-01_T1w.json
 	│	└── sub-01_ses-01_T1w.nii.gz
 	├── dwi
 	│	├── sub-01_ses-01_acq-b2000_dir-AP_dwi.bval

```

--------------------------------

### Directory Structure Example

Source: https://micapipe.readthedocs.io/en/latest/pages/04.matrices/index.html

Illustrates the directory structure of pipeline outputs, showing the organization of subject-specific data and connectome files.

```text
out
├── fastsurfer
└── micapipe_v0.2.0


```

```text
micapipe_v0.2.0/
├── micapipe_processed_sub.csv
├── dataset_description.json
├── micapipe_group-QC.pdf
├── sub-HC001
├── sub-HC002
├── sub-HC003
...


```

```text
sub-HC001/
└── ses-01
    ├── **anat**
    ├── **dist**
    ├── **dwi**
    │   ├── **connectomes**      # DWI connectomes
    │   └── eddy                 # fsl eddy outputs
    ├── **func**
    │   ├── desc-me_task-rest_bold
    │   │   ├── surf
    │   │   └── volumetric
    │   └── desc-me_task-semantic_bold
    │       ├── surf
    │       └── volumetric
    ├── logs                     # log files
    ├── **maps**
    ├── **mpc**
    ├── QC
    │   └── eddy_QC              # fsl eddy_quad outputs
    ├── **surf**
    └── xfm                      # Transformation matrices and warpfields

```

--------------------------------

### Clone micapipe repository

Source: https://micapipe.readthedocs.io/en/latest/pages/01.install/index.html

Download the micapipe source code from GitHub to set up a bare-metal installation.

```bash
$ git clone https://github.com/MICA-LAB/micapipe.git

```

--------------------------------

### Example BIDS Directory Structure (Initial)

Source: https://micapipe.readthedocs.io/en/latest/pages/05.mic2bids/index.html

Shows the basic BIDS directory layout for a subject and session, indicating where different types of imaging data will be stored.

```text
sub-01
└── ses-01
 	├── anat
 	├── dwi
 	└── func

```

--------------------------------

### R Environment Setup for Connectome Visualization

Source: https://micapipe.readthedocs.io/en/latest/pages/04.matrices/index.html

Sets up the R environment by loading necessary packages, defining the working directory, subject identifiers, and the atlas for connectome visualization.

```r
# Set the environment
require("RColorBrewer")
require("viridis")
require("gifti")

# Set the working directory to your subject's directory
setwd("out/micapipe_v0.2.0/sub-HC001/ses-01")

# This variable will be different for each subject
subjectID <- 'sub-HC001_ses-01'

# Here we define the atlas
atlas <- 'schaefer-400'


```

--------------------------------

### Add ANTs dependency to PATH

Source: https://micapipe.readthedocs.io/en/latest/pages/01.install/index.html

Include the ANTs installation directory in the system's PATH for bare-metal installations. This is an example, and other dependencies must also be added similarly.

```bash
$ export ANTSDIR="/Path/To/ANTs"
$ PATH=${PATH}:${ANTSDIR}
$ export PATH

```

--------------------------------

### Python Environment Setup for Connectome Visualization

Source: https://micapipe.readthedocs.io/en/latest/pages/04.matrices/index.html

Sets up the Python environment by defining paths, subject identifiers, and the atlas for loading and visualizing connectome matrices.

```python
# Load required packages
import os
import numpy as np
import nibabel as nib
from nilearn import plotting
import matplotlib as plt

# Set the working directory to the 'out' directory
os.chdir("/data_/mica3/BIDS_MICs/derivatives") # <<<<<<<<<<<< CHANGE THIS PATH TO YOUR OUT DIRECTORY

# This variable will be different for each subject
sub='HC001'           # <<<<<<<<<<<< CHANGE THIS SUBJECT's ID
ses='01'              # <<<<<<<<<<<< CHANGE THIS SESSION
subjectID=f'sub-{sub}_ses-{ses}'
subjectDir=f'micapipe_v0.2.0/sub-{sub}/ses-{ses}'

# Here we define the atlas
atlas='schaefer-400'


```

--------------------------------

### Set micapipe environment variables

Source: https://micapipe.readthedocs.io/en/latest/pages/01.install/index.html

Add the micapipe installation directory and its functions directory to the system's PATH for bare-metal usage.

```bash
$ export MICAPIPE=/Path/To/Cloned/Micapipe/Repo
$ PATH=${PATH}:${MICAPIPE}:${MICAPIPE}/functions
$ export PATH

```

--------------------------------

### Visualize Surface Data

Source: https://micapipe.readthedocs.io/en/latest/pages/04.surfaces/index.html

Example of visualizing surface data using a Python script. Ensure necessary libraries are imported.

```python
import nibabel as nib
from nilearn import surface

# Load surface data (e.g., from a GIFTI file)
surface_data = nib.load('path/to/your/surface_data.shape.gii')

# Visualize the surface data
surface.view(surface_data)
```

--------------------------------

### Smooth surfaces using WorkBench's wb_command

Source: https://micapipe.readthedocs.io/en/latest/pages/04.faq/index.html

Smooth surface data using WorkBench's wb_command, which requires converting MGH files to GIFTI format first. This example demonstrates smoothing the left hemisphere surface with a 20mm FWHM.

```bash
MICAPIPE_DIR=<path to the micapipe repository>

# For WorkBench the first step is to convent the mgh surface file to GIFTI
mri_convert "${outDir}/sub-01_ses01_space-conte69-32k_desc-lh_thickness.mgh" "/tmp/lh_curv_c69-32k_thickness.func.gii"

wb_command -metric-smoothing \
    "${MICAPIPE_DIR}/surfaces/fsaverage.L.midthickness_orig.32k_fs_LR.surf.gii" \
    "/tmp/lh_curv_c69-32k_thickness.func.gii" \
    20 \
    "/tmp/lh_curv_20mm_c69-32k.func.gii"     # This is the 20 mm FWHM surface

# Convert from GIFTI back to MGH
mri_convert "/tmp/lh_curv_10mm_c69-32k.func.gii" "${outDir}/sub-01_ses01_space-conte69-32k_desc-lh_thickness_20mm.mgh"
```

--------------------------------

### Micapipe Processing Command for Audiopath Dataset

Source: https://micapipe.readthedocs.io/en/latest/pages/04.databases/index.html

Example command to process the Audiopath dataset using micapipe. This command specifies structural, surface, diffusion, and functional MRI processing steps, including options for reverse phase encoding and microstructural imaging.

```bash
1 subjectDir=./ds001942/sub-S01/ses-SES01
2 dwiDir=./ds001942/sub-S01/ses-SES01/dwi
3 sub=sub-S02_ses-SES01
4
5 micapipe -bids ds001942/ -out . -sub S01 -ses SES01 \
6           -proc_structural -t1wStr run-01_T1w,run-02_T1w \
7           -proc_surf \
8           -post_structural \
9           -proc_dwi \
10             -dwi_main ${dwiDir}/${sub}_dir-AP_run-01_dwi.nii.gz,${dwiDir}/${sub}_dir-AP_run-02_dwi.nii.gz,${dwiDir}/${sub}_dir-AP_run-03_dwi.nii.gz \
11             -dwi_rpe ${dwiDir}/${sub}_dir-PA_run-01_dwi.nii.gz,${dwiDir}/${sub}_dir-PA_run-02_dwi.nii.gz,${dwiDir}/${sub}_dir-PA_run-03_dwi.nii.gz \
12             -rpe_all \
13           -SC \
14           -proc_func \
15             -noFIX -NSR \
16             -mainScanStr task-rest_bold \
17             -func_pe  ${subjectDir}/fmap/${sub}_acq-gre_dir-AP_run-01_epi.nii.gz \
18             -func_rpe ${subjectDir}/fmap/${sub}_acq-gre_dir-PA_run-01_epi.nii.gz \
19           -GD \
20           -MPC \
21             -microstructural_img ${subjectDir}/anat/${sub}_acq-ShortInv_run-01_T1w.nii.gz \
22             -microstructural_reg ${subjectDir}/anat/${sub}_acq-ShortInv_run-01_T1w.nii.gz \
23           -QC_subj
```

--------------------------------

### Set Up Environment and Subject Directories

Source: https://micapipe.readthedocs.io/en/latest/pages/04.fetch/index.html

Configures the environment by setting the working directory and defining subject-specific paths. Remember to update the paths and subject/session IDs.

```python
# Set the environment
import os
import glob
import numpy as np
import nibabel as nib
from brainspace.plotting import plot_hemispheres
from brainspace.mesh.mesh_io import read_surface
from brainspace.datasets import load_conte69
from brainspace.gradient import GradientMaps
from brainspace.utils.parcellation import map_to_labels
import matplotlib.pyplot as plt
import cmocean

# Add cmocean maps to cmaps variable
cmaps = cmocean.cm.cmap_d

# Set the working directory to the 'out' directory
out='/data_/mica3/BIDS_PNI/derivatives'
os.chdir(out)     # <<<<<<<<<<<< CHANGE THIS PATH

# This variable will be different for each subject
sub='PNC019' # <<<<<<<<<<<< CHANGE THIS SUBJECT's ID
ses='a1'    # <<<<<<<<<<<< CHANGE THIS SUBJECT's SESSION
subjectID=f'sub-{sub}_ses-{ses}'
subjectDir=f'micapipe_v0.2.0/sub-{sub}/ses-{ses}'

# Path to MICAPIPE from global enviroment
micapipe=os.popen("echo $MICAPIPE").read()[:-1] # <<<<<<<<<<<< CHANGE THIS PATH

```

--------------------------------

### Set Up Python Environment and Variables

Source: https://micapipe.readthedocs.io/en/latest/pages/04.gradients/index.html

Initializes the Python environment by importing necessary libraries and setting up subject-specific paths and parameters for gradient analysis.

```python
 1# Set the environment
 2import os
 3import glob
 4import numpy as np
 5import nibabel as nib
 6from brainspace.plotting import plot_hemispheres
 7from brainspace.mesh.mesh_io import read_surface
 8from brainspace.datasets import load_conte69
 9from brainspace.gradient import GradientMaps
10from brainspace.utils.parcellation import map_to_labels
11import matplotlib.pyplot as plt
12
13# Set the working directory to the 'out' directory
14out='/data_/mica3/BIDS_MICs/derivatives'
15os.chdir(out)     # <<<<<<<<<<<< CHANGE THIS PATH
16
17# This variable will be different for each subject
18sub='HC001' # <<<<<<<<<<<< CHANGE THIS SUBJECT's ID
19ses='01'    # <<<<<<<<<<<< CHANGE THIS SUBJECT's SESSION
20subjectID=f'sub-{sub}_ses-{ses}'
21subjectDir=f'micapipe_v0.2.0/sub-{sub}/ses-{ses}'
22
23# Here we define the atlas
24atlas='schaefer-400' # <<<<<<<<<<<< CHANGE THIS ATLAS
25
26# Path to MICAPIPE from global enviroment
27micapipe=os.popen("echo $MICAPIPE").read()[:-1] # <<<<<<<<<<<< CHANGE THIS PATH

```

--------------------------------

### Set Environment - Python

Source: https://micapipe.readthedocs.io/en/latest/pages/04.surfaces/index.html

Sets up the Python environment by importing necessary libraries and defining working directories and subject-specific paths. Ensure the 'out' path points to your derivatives directory.

```python
# Set the environment
import os
import glob
import numpy as np
import nibabel as nib
import seaborn as sns
from brainspace.plotting import plot_hemispheres
from brainspace.mesh.mesh_io import read_surface
from brainspace.datasets import load_conte69

# Set the working directory to the 'out' directory
out='/data_/mica3/BIDS_MICs/derivatives' # <<<<<<<<<<<< CHANGE THIS PATH
os.chdir(out)

# This variable will be different for each subject
sub='sub-HC001'
ses='ses-01'
subjectID=f'{sub}_{ses}'           # <<<<<<<<<<<< CHANGE THIS SUBJECT's ID
subjectDir=f'micapipe_v0.2.0/{sub}/{ses}' # <<<<<<<<<<<< CHANGE THIS SUBJECT's DIRECTORY

# Set paths and variables
dir_FS = 'freesurfer/' + subjectID
dir_surf = subjectDir + '/surf/'
dir_maps = subjectDir + '/maps/'

# Path to MICAPIPE
micapipe=os.popen("echo $MICAPIPE").read()[:-1]

```

--------------------------------

### Use FreeSurfer for Surface Reconstruction

Source: https://micapipe.readthedocs.io/en/latest/pages/02.structuralproc/index.html

Specify the -freesurfer option with the path to your FreeSurfer installation to use it instead of the default FastSurfer for surface reconstruction.

```bash
-freesurfer <path>
```

--------------------------------

### Build Singularity Image (>= 2.5)

Source: https://micapipe.readthedocs.io/en/latest/pages/01.install/index.html

Create a Singularity image directly on an HPC using the 'singularity build' command. Replace '<version>' with the desired micapipe version.

```bash
$ singularity build /my_images/micapipe-<version>.simg docker://micalab/micapipe:<version>

```

```bash
$ singularity build /my_images/micapipe-v0.2.3.simg docker://micalab/micapipe:v0.2.3

```

--------------------------------

### Downsample Tractogram Example

Source: https://micapipe.readthedocs.io/en/latest/pages/02.flairproc/index.html

Demonstrates how to downsample a tractogram using Python. This is useful for reducing the size of tractogram files while preserving essential information.

```python
from micapipe.pipelines.flairproc.utils.tractogram_utils import downsample_tractogram

# Example usage:
downsample_tractogram(
    input_tractogram="path/to/your/tractogram.trk",
    output_tractogram="path/to/your/downsampled_tractogram.trk",
    target_nb_points=1000000,
    seed=42
)
```

--------------------------------

### Create Singularity Image (< 2.5) on Windows

Source: https://micapipe.readthedocs.io/en/latest/pages/01.install/index.html

Adapt the 'docker2singularity' command for Windows systems by adjusting the volume mount for the output directory. Ensure the path format is correct for Windows.

```bash
$ docker run --privileged -t --rm \
    -v /var/run/docker.sock:/var/run/docker.sock \
    -v D:\host\path\where\to\output\singularity\image:/output \
    singularityware/docker2singularity \
    micalab/micapipe:v0.2.3

```

--------------------------------

### Example DICOM Directory Structure

Source: https://micapipe.readthedocs.io/en/latest/pages/05.mic2bids/index.html

Illustrates the typical organization of DICOM files within a sorted directory, where subdirectories correspond to different MRI sequences.

```text
S1_02_AAHead_Scout_64ch-head-coil_MPR_sag
S1_03_AAHead_Scout_64ch-head-coil_MPR_cor
S1_04_AAHead_Scout_64ch-head-coil_MPR_tra
S1_08_mprage-0.8iso_ORIG
S1_10_dwi_b2000_90
S1_11_dwi_b700_40
S1_12_dwi_b300_10
S1_13_dwi_b0_5PA
S1_16_rsfmri-3mm-bold_AP
S1_17_rsfmri-3mm_se_AP
S1_18_rsfmri-3mm_se_PA
S1_20_mprage-0.8iso_ORIG
S1_29_mp2rage-0.8iso-sess2_INV1
S1_30_mp2rage-0.8iso-sess2_T1_Images
S1_31_mp2rage-0.8iso-sess2_UNI_Images
S1_32_mp2rage-0.8iso-sess2_INV2

```

--------------------------------

### Set Environment - R

Source: https://micapipe.readthedocs.io/en/latest/pages/04.surfaces/index.html

Sets up the R environment by loading required libraries and defining working directories and subject-specific paths. Ensure the 'setwd' path is correctly set.

```r
# Set the environment        'R 4.3.1'
require('RColorBrewer')      # version 1.1.3
require('viridis')           # version 0.6.5
require('fsbrain')           # version 0.5.5
require('freesurferformats') # version 0.1.18
require('rgl')               # version 1.2.8
require('gifti')             # version 0.8.0

# Define working directory and subject-specific information. CHANGE THESE      PATHS as appropriate.
setwd('~/derivatives_RtD/')                      # working directory
subjectID <- 'sub-HC001_ses-01'                  # subject ID
subjectDir <- 'micapipe_v0.2.0/sub-HC001/ses-01' # subject directory path

# Set paths for surface and morphometry data.
dir_surf <- paste0(subjectDir, '/surf/')
dir_maps <- paste0(subjectDir, '/maps/')


```

--------------------------------

### Display Micapipe Help Information

Source: https://micapipe.readthedocs.io/en/latest/pages/01.whatyouneed/index.html

Use the -help or -h flag to display a list and description of all available arguments and flags for Micapipe.

```bash
micapipe -help
```

```bash
micapipe -h
```

--------------------------------

### Run -proc_flair module

Source: https://micapipe.readthedocs.io/en/latest/pages/02.flairproc/index.html

This command initiates the FLAIR preprocessing pipeline. Ensure you have run the structural and surface preprocessing steps beforehand. Replace placeholders with your specific BIDS directory, subject ID, and output directory.

```bash
$ micapipe **-bids** <BIDS-directory> **-sub** <subject_id> **-out** <outputDirectory> **-proc_flair** <options>
```

--------------------------------

### Python Code Example for Tractogram Downsampling

Source: https://micapipe.readthedocs.io/en/latest/pages/05.mic2bids/index.html

Demonstrates how to downsample a tractogram using Python. This snippet is useful for reducing the number of points or streamlines lengths in a tractogram.

```python
import micapipe.utils.datasets as datasets

# Example of downsampling a tractogram
# This is a placeholder for the actual code example which would be
# provided in a real documentation scenario.
# For instance, it might involve loading a tractogram, specifying
# downsampling parameters, and saving the result.

# Placeholder for tractogram downsampling logic
print("Tractogram downsampling example would go here.")

# Example of accessing dataset information (if relevant to downsampling context)
# datasets.get_dataset_info('some_dataset_id')

```

--------------------------------

### Tract Density Image (TDI) Example Path

Source: https://micapipe.readthedocs.io/en/latest/pages/02.dwiproc/index.html

Shows the file path for the generated Tract Density Image (TDI), which is used for assessing DWI processing quality.

```bash
File: <outputDirectory>/micapipe_v0.2.0/<subject_id>/dwi/<subject_id>_space-dwi_desc-iFOD2-<tracts>_tdi.mif
```

--------------------------------

### Running the mic2bids Script

Source: https://micapipe.readthedocs.io/en/latest/pages/05.mic2bids/index.html

Execute the mic2bids script with input and output directories, subject ID, and session information. The -force flag can be used to overwrite existing data.

```bash
mic2bids  -in <DICOMS_directory> -bids <BIDS directory path> -id <subject> -ses <session>
```

--------------------------------

### Select Specific T1w Scan for Processing

Source: https://micapipe.readthedocs.io/en/latest/pages/02.structuralproc/index.html

Use the -T1wStr option to specify which T1w scan to process when multiple are present. This example processes the scan named 'run-02_T1w'.

```bash
1 micapipe -sub <subject_id> -out <outputDirectory> -bids <BIDS-directory> \
2           -proc_structural -T1wStr run-02_T1w

```

--------------------------------

### Run -MPC_SWM command

Source: https://micapipe.readthedocs.io/en/latest/pages/02.mpcswm/index.html

This is the basic command to run the -MPC_SWM module. Ensure you have the necessary prerequisites and provide the subject ID, output directory, and BIDS directory.

```bash
micapipe **-sub** <subject_id> **-out** <outputDirectory> **-bids** <BIDS-directory> **-MPC_SWM** <options>
```

--------------------------------

### Create Singularity Image (< 2.5) using Docker

Source: https://micapipe.readthedocs.io/en/latest/pages/01.install/index.html

Generate a Singularity image from a Docker image using 'docker2singularity' on a system with Docker installed. Specify the output path for the image.

```bash
$ docker run --privileged -t --rm \
    -v /var/run/docker.sock:/var/run/docker.sock \
    -v /absolute/path/to/output/folder:/output \
    singularityware/docker2singularity \
    micalab/micapipe:<version>

```

```bash
$ docker run --privileged -t --rm \
    -v /var/run/docker.sock:/var/run/docker.sock \
    -v /absolute/path/to/output/folder:/output \
    singularityware/docker2singularity \
    micalab/micapipe:v0.2.3

```

--------------------------------

### Run Structural Pre-processing

Source: https://micapipe.readthedocs.io/en/latest/pages/02.structuralproc/index.html

This command initiates the structural pre-processing pipeline. It requires subject ID, output directory, and BIDS directory as mandatory arguments.

```bash
$ micapipe **-sub** <subject_id> **-out** <outputDirectory> **-bids** <BIDS-directory> **-proc_structural** **<options>**
```

--------------------------------

### GD Connectomes fsLR-32k Example

Source: https://micapipe.readthedocs.io/en/latest/pages/04.fetch/index.html

Loads and plots Geodesic Distance (GD) connectomes for the fsLR-32k surface using multiple parcellations. It visualizes the connectome row sums across different atlases.

```python
# Parcellations to load
parc = ['aparc', 'economo', 'glasser-360', 'schaefer-400', 'vosdewael-400']
N = len(parc)

# Load the connectome row sum for each parcellation
roi_fs5, roi_32k = load_connectomes('GD', parc, subjectDir, subjectID)

# Plot parcellations
plot_hemispheres(f32k_lh, f32k_rh, array_name=roi_32k[0:N], size=(600, 88*N), cmap='RdBu_r',
                 embed_nb=True, label_text={'left':parc[0:N]}, color_bar='right',
                 zoom=1.5, nan_color=nan_col)
```

--------------------------------

### Create and Display fsLR-5k Surface with Thickness Data (R)

Source: https://micapipe.readthedocs.io/en/latest/pages/04.surfaces/index.html

Loads fsLR-5k surface geometry and thickness data, creates colored meshes, and displays the surface. Ensure `dir_maps` and `subjectID` are defined.

```R
# Load fsLR-5k surface paths for left and right hemispheres.
f5k.lh <-  read.fs.surface(filepath = 'micapipe/surfaces/fsLR-5k.L.surf.gii', format = "gii")
f5k.rh <-  read.fs.surface(filepath = 'micapipe/surfaces/fsLR-5k.R.surf.gii', format = "gii")

# Define paths for thickness morphometry data on fsLR-5k for each hemisphere.
f5k.lh.th <- paste0(dir_maps, subjectID, '_hemi-L_surf-fsLR-5k_label-thickness.func.gii')
f5k.rh.th <- paste0(dir_maps, subjectID, '_hemi-R_surf-fsLR-5k_label-thickness.func.gii')

# Create colored meshes based on thickness data.
cml.f5k.th <- coloredmesh.from.preloaded.data(f5k.lh, morph_data = lft(unlist(readgii(f5k.lh.th)$data)), hemi = 'lh', makecmap_options = list('colFn'=inferno))
cmr.f5k.th <- coloredmesh.from.preloaded.data(f5k.rh, morph_data = lft(unlist(readgii(f5k.rh.th)$data)), hemi = 'rh', makecmap_options = list('colFn'=inferno))
th_f5k <- brainviews(views = 't4', coloredmeshes=list('lh'=cml.f5k.th, 'rh'=cmr.f5k.th), rglactions = list('no_vis'=T))

# Display the fsLR-5k surface with thickness data.
plot_surface(th_f5k, 'Thickness [mm]')
```

--------------------------------

### Register Selected Parcellations with -post_structural

Source: https://micapipe.readthedocs.io/en/latest/pages/02.structuralproc/index.html

This example demonstrates how to use the -atlas argument with the -post_structural module to register only specific parcellations (schaefer-200, economo, aparc) to the subject's space. This is useful for focusing subsequent analyses.

```bash
micapipe -sub <subject_id> -out <outputDirectory> -bids <BIDS-directory> \
           -post_structural -atlas schaefer-200,economo,aparc
```

--------------------------------

### Downsample Tractogram with Python

Source: https://micapipe.readthedocs.io/en/latest/pages/04.tckdownsample/index.html

This Python function compresses a tractogram file by reducing the number of points on each streamline using `approx_polygon_track`. It loads the tractogram, applies the downsampling, and saves the new compressed file. Ensure DIPY and nibabel are installed.

```python
import nibabel as nib
from dipy.io.streamline import load_tractogram, save_tractogram
from dipy.tracking.distances import approx_polygon_track

def compress_tck(file_name):

    # Load reference anatomy
    reference_anatomy=nib.load(subject + '/dwi/' + subject + '_ses-01_space-dwi_desc-b0.nii.gz')

    # load tractogram
    bundle = load_tractogram(file_name, reference_anatomy, bbox_valid_check=False)
    print('This bundle has %d streamlines' % len(bundle))

    #tractogram size reduction
    bundle_downsampled = [approx_polygon_track(s, 0.25) for s in bundle.streamlines]

    # Replace the tractogram with the downsample
    bundle.streamlines = bundle_downsampled

    # Save new TCK file
    save_tractogram(bundle, file_name.replace('.tck','_downsampled.tck'))
    print('Downsampled tractogram was saved')

# Compress a tractogram
subject = 'sub-01'
tck = subject + '/dwi/' + subject + '_space-dwi_desc-iFOD1-10M_tractography.tck'
compress_tck(tck)
```

--------------------------------

### Run -post_structural Module

Source: https://micapipe.readthedocs.io/en/latest/pages/02.structuralproc/index.html

This command initiates the structural post-processing stage. It requires subject ID, output directory, BIDS directory, and the -post_structural flag. Optional arguments can specify which parcellations to register.

```bash
micapipe **-sub** <subject_id> **-out** <outputDirectory> **-bids** <BIDS-directory> **-post_structural** **<options>**
```

--------------------------------

### Micapipe Processing Command for Cam-CAN Data

Source: https://micapipe.readthedocs.io/en/latest/pages/04.databases/index.html

Example command to process Cam-CAN data using micapipe, including structural, diffusion, and functional MRI pipelines. It specifies output directories, subject/session IDs, and various processing flags.

```bash
 1 micapipe -bids rawdata -out derivatives -sub CC110045 -ses pre \
 2           -proc_structural \
 3           -proc_surf \
 4             -surf_dir ./freesurfer_processed_CC110045 \
 5             -freesurfer \
 6           -post_structural \
 7           -proc_dwi \
 8           -SC \
 9           -proc_func \
10             -mainScanStr task-Rest_bold \
11             -NSR \
12             -noFIX \
13           -GD \
14           -MPC \
15             -microstructural_img sub-CC110045/ses-pre/mti_MTR.nii \
16           -QC_subj \

```

--------------------------------

### Run Surface Reconstruction

Source: https://micapipe.readthedocs.io/en/latest/pages/02.structuralproc/index.html

This command initiates the surface reconstruction pipeline using the -proc_surf module. Specify subject ID, output directory, BIDS directory, and any relevant options.

```bash
$ micapipe **-sub** <subject_id> **-out** <outputDirectory> **-bids** <BIDS-directory> **-proc_surf** **<options>**
```

--------------------------------

### Create and Display fsLR-32k Surface with Thickness Data (R)

Source: https://micapipe.readthedocs.io/en/latest/pages/04.surfaces/index.html

Loads fsLR-32k surface geometry and thickness data, creates colored meshes, and displays the surface. Requires `dir_maps` and `subjectID` to be defined.

```R
# Load fsLR-32k surface paths for left and right hemispheres.
f32k.lh <-  read.fs.surface(filepath = 'micapipe/surfaces/fsLR-32k.L.surf.gii', format = "gii")
f32k.rh <-  read.fs.surface(filepath = 'micapipe/surfaces/fsLR-32k.R.surf.gii', format = "gii")

# Define paths for thickness morphometry data on fsLR-32k for each hemisphere.
f32k.lh.th <- paste0(dir_maps, subjectID, '_hemi-L_surf-fsLR-32k_label-thickness.func.gii')
f32k.rh.th <- paste0(dir_maps, subjectID, '_hemi-R_surf-fsLR-32k_label-thickness.func.gii')

# Create colored meshes based on thickness data.
cml.f32k.th <- coloredmesh.from.preloaded.data(f32k.lh, morph_data = lft(unlist(readgii(f32k.lh.th)$data)), hemi = 'lh', makecmap_options = list('colFn'=inferno))
cmr.f32k.th <- coloredmesh.from.preloaded.data(f32k.rh, morph_data = lft(unlist(readgii(f32k.rh.th)$data)), hemi = 'rh', makecmap_options = list('colFn'=inferno))
th_f32k <- brainviews(views = 't4', coloredmeshes=list('lh'=cml.f32k.th, 'rh'=cmr.f32k.th), rglactions = list('no_vis'=T))

# Display the fsLR-32k surface with thickness data.
plot_surface(th_f32k, 'Thickness [mm]')
```

--------------------------------

### Register Functional Sessions to a Specific Anatomical Session

Source: https://micapipe.readthedocs.io/en/latest/pages/02.restingstateproc/index.html

When processing longitudinal data with multiple sessions, use the -sesAnat flag to register all functional sessions to a common anatomical reference. This ensures consistency across sessions for analysis. The example shows registering 'func01' and 'func02' to the 'struct01' anatomical session.

```bash
1 micapipe -sub 01 -ses func01 -proc_func -sesAnat struct01 -bids <bidsDir> -out <outDir>
2 micapipe -sub 01 -ses func02 -proc_func -sesAnat struct01 -bids <bidsDir> -out <outDir>

```

--------------------------------

### Run micapipe with Docker

Source: https://micapipe.readthedocs.io/en/latest/pages/01.install/index.html

Execute micapipe within a Docker container, mounting necessary directories and specifying processing parameters. Ensure correct permissions for mounted directories.

```bash
 1bids=<path to bids dataset>
 2out=<path to outputs>
 3fs_lic=<path to freesurfer licence>
 4docker run -ti --rm \
 5    -v ${bids}:/bids \
 6    -v ${out}:/out \
 7    -v ${tmp}:/tmp \
 8    -v ${fs_lic}:/opt/licence.txt \
 9    micalab/micapipe:v0.2.3 \
10        -bids /bids -out /out -fs_licence /opt/licence.txt -threads 10 \
11        -sub $sub -ses ${ses} \
12        -proc_structural

```

--------------------------------

### Load and Plot Intensity Profiles (Python)

Source: https://micapipe.readthedocs.io/en/latest/pages/04.matrices/index.html

Loads intensity profiles from a .shape.gii file and plots them as a matrix. Requires the nibabel and nilearn libraries.

```python
# Set the path to the Intensity profiles file
cnt_int = subjectDir + f'/mpc/{mpc_acq}/{subjectID}_atlas-{atlas}_desc-intensity_profiles.shape.gii'

# Load the Intensity profiles
mtx_int = nib.load(cnt_int).darrays[0].data

# Plot as a matrix
corr_plot = plotting.plot_matrix(mtx_int, figure=(20,10), labels=None, cmap='Greens', colorbar=False)

```

--------------------------------

### Create and Display fs5 Surface with Thickness Data (R)

Source: https://micapipe.readthedocs.io/en/latest/pages/04.surfaces/index.html

Generates a colored mesh for the fs5 surface based on thickness data and displays it. Requires preloaded fs5 surface data and thickness data files.

```R
cml.fs5.th <- coloredmesh.from.preloaded.data(fs5.lh, morph_data = lft(unlist(readgii(fs5.lh.th)$data)), hemi = 'lh', makecmap_options = list('colFn'=inferno))
cmr.fs5.th <- coloredmesh.from.preloaded.data(fs5.rh, morph_data = lft(unlist(readgii(fs5.rh.th)$data)), hemi = 'rh', makecmap_options = list('colFn'=inferno))
th_fs5 <- brainviews(views = 't4', coloredmeshes=list('lh'=cml.fs5.th, 'rh'=cmr.fs5.th), rglactions = list('no_vis'=T))

# Display the fsaverage5 surface with thickness data.
plot_surface(th_fs5, 'Thickness [mm]')
```

--------------------------------

### Load qMRI Surface Data (Python)

Source: https://micapipe.readthedocs.io/en/latest/pages/04.surfaces/index.html

Loads qMRI intensity maps from a specified surface. Ensure the 'dir_maps' and 'dir_surf' variables are set, and necessary libraries like 'glob', 'nib', and 'numpy' are imported.

```python
import glob
import nibabel as nib
import numpy as np

# Assume dir_maps and dir_surf are defined globally or passed as arguments
# dir_maps = "/path/to/maps/"
# dir_surf = "/path/to/surfaces/"

def load_qmri(qmri='', surf='fsLR-32k'):
    """
    This function loads the qMRI intensity maps from midthickness surface
    """
    # List the files
    files_lh = sorted(glob.glob(f"{dir_maps}/*_hemi-L_surf-{surf}_label-midthickness_{qmri}.func.gii"))
    files_rh = sorted(glob.glob(f"{dir_maps}/*_hemi-R_surf-{surf}_label-midthickness_{qmri}.func.gii"))

    # Load map data
    surf_map=np.concatenate((nib.load(files_lh[0]).darrays[0].data, nib.load(files_rh[0]).darrays[0].data), axis=0)

    return(surf_map)

def plot_qmri(qmri='',  surf='fsLR-32k', label='pial', cmap='rocket', rq=(0.15, 0.95)):
    """
    This function plots the qMRI intensity maps on the pial surface
    """
    # Load the data
    map_surf = load_qmri(qmri, surf)
    print('Number of vertices: ' + str(map_surf.shape[0]))

    # Load the surfaces
    surf_lh=read_surface(f'{dir_surf}/{subjectID}_hemi-L_space-nativepro_surf-{surf}_label-{label}.surf.gii', itype='gii')
    surf_rh=read_surface(f'{dir_surf}/{subjectID}_hemi-R_space-nativepro_surf-{surf}_label-{label}.surf.gii', itype='gii')

    # Color range based in the quantiles
    crange=(np.quantile(map_surf, rq[0]), np.quantile(map_surf, rq[1]))

    # Plot the group T1map intensitites
    fig = plot_hemispheres(surf_lh, surf_rh, array_name=map_surf, size=(900, 250), color_bar='bottom', zoom=1.25, embed_nb=True, interactive=False, share='both',
                     nan_color=(0, 0, 0, 1), cmap=cmap, color_range=crange, transparent_bg=False, screenshot = False)
    return(fig)
```
Tokens:
Raw