forked from Washington-University/HCPpipelines
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathhrb_pipelines_outline.txt
85 lines (65 loc) · 4.58 KB
/
hrb_pipelines_outline.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
HCP Preprocessing Pipelines: Notes
Holland Brown
Updated 2023-02-08
Created 2023-02-08
See HCP Pipelines Installation and Usage Instructions:
https://github.com/Washington-University/HCPpipelines/wiki/Installation-and-Usage-Instructions
---------------------------------------------------------------------
- Liston ME fMRI Preprocessing Pipeline scripts predicated on these pipelines
- before running, see example file structure required
Preprocessing -------------------------------------------------------
1. Structural Preprocessing Pipeline
1. PreFreeSurfer.sh
- see script: ${HCPPIPEDIR}/Examples/Scripts/PreFreeSurferPipelineBatch.sh
- it runs a batch of subj data through this part of the pipeline
1. edit StudyFolder vbl --> path to folder that will contain all subjs' data within your login/home dir
- example StudyFolder dir: "${HOME}/projects/Pipelines_ExampleData"
2. edit Subjlist vbl >> space delimited list of subj ids to be run by PreFreeSurfer
- can also submit a string with one subj id if you need to run only one subj
- script will look for subj id folders in specified StudyFolder dir for data to run
3. set EnvironmentScript vbl >> script that sets up env vbls
- see example env setup: Examples/Scripts/SetUpHCPPipeline.sh
- update script to reflect where HCP Pipelines installed
4. set QUEUE vbl to reflect proc queue if submitting job to scheduler
- queue names are determined by cluster setup
- leave vbl empty to run locally
5. Gradient Distortion Coefficients
- set to contain either path to gradient coeffs file or "NONE" to skip gradient distortion correction
2. FreeSurfer.sh
3. PostFreeSurfer.sh
2. Functional Preprocessing Pipeline (after struct preproc)
- dirs are created for task ana during this step, so know how tasks will be modeled/analyzed
1. Generic fMRI volume preprocessing (always do this step before surface preproc)
2. Generic fMRI surface preprocessing
Optional Processing ---------------------------------------------------
3. ICA FIX Pipeline Pipeline (after functional preprocessing)
- ICA-based denoising pipline
- regresses out motion timeseries and artifact independent components
4. MSAII Pipeline (after functional preprocessing and ICA FIX processing)
- re-registration of cortical surfaces
- uses cortical folding, myelin map, and rsfMRI data
Task Analysis, i.e. tfMRI Processing -------------------------------------
5. Preparing for Level 1 tfMRI Analysis (after functional preprocessing)
- also recommended to do ICA FIX and temporal ICA (tICA) before this step
- Level 1 is for within-subs, within-runs ana
1. Level 1 FEAT setup (FSF) files for ea direction of fnctl task
- use templates in HCP directory: ${HCPPIPEDIR}/Examples/fsf_templates
- num time pts in template must match num time pts in corresponding scan
- see Mumford FSL Tutorial notes for search-and-replace script to create FSF files
- copy template for a scan into: MNINonLinear/Results/<task>
- make sure num time pts matches num volumes for the scan
- can use script: ${HCPPIPEDIR}/Examples/Scripts/generate_level1_fsf.sh to extract num time pts from img file and set num vols in FSF file
- must do this for every task direction so there is an FSF file for each
2. E-Prime EV files in MNINonLinear/Results subdir for ea task
- this subdirectory is created during Functional Preprocessing Pipeline, so can't do this until after that
- each task has its own dir, so must do this for each
- also, this means I have to understand how the tasks will be modeled before functional preprocessing
- use script: ${HCPPIPEDIR}/Examples/Scripts/copy_evs_into_results.sh to copy EV files to this directory
6. Preparing for Level 2 tfMRI Analysis (after functional preprocessing)
1. Level 2 FEAT setup files for ea task direction for which there will be level 2 ana
- must be in dir: <StudyFolder>/100307/MNINonLinear/Results/<task>/<subj><task>.fsf
- can copy template: ${HCPPIPEDIR}/Examples/fsf_templates/tfMRI_EMOTION_hp200_s4_level2.fsf to run example analysis
- likely will have to create level 2 results directory: <StudyFolder>/100307/MNINonLinear/Results/<Study Name>
Diffusion Preprocessing -----------------------------------------------------------
- requires structural preprocessing already complete
- see script: ${HCPPIPEDIR}/Examples/Scripts/DiffusionPreprocessingBatch.sh is similar to struct preprocessing scripts