NITRC C-PAC Forum: c-pac

Questions regarding C-PAC nuisance signal regression

mihai_avram — Mon, 07 Mar 2022 7:34:11 GMT

Hi everyone,

I hope someone can help me with two issues regarding NSR in C-PAC.

(a) The first once concerns the order of the NSR steps implemented in the default C-PAC pipeline. I think all the desired steps (e.g. aCompCor, motion correction, GSR etc.) are performed simultaneously and not in a modular/step-wise fashion but can anyone confirm this is correct? An additional step that can be performed after or before regression is the band-pass filtering.

(b) The second issue concerns global signal regression. As CSF and WM contain some variance related to GSR, it appears possible that CompCor may actually account for some correction similar to GSR. This would of course be related to how the CSF and WM look like, and it seems that the default pipeline includes erosion (thus minimizing a possible effect). But it's unclear to me what the level of erosion is (e.g., 30mm?). For instance, there is some evidence that eroding the WM twice and CSF four times removes the risk of unwilling correcting for global signal (Power et al., 2017: https://doi.org/10.1016/j.neuroimage.2016.09.038). I'm trying to avoid the possibility of regressing the global signal "twice", once via CompCor and once via GSR, as I have both active in my pipeline. Is there a risk of this? The CSF and WM masks used in the NSR also appear to have a high threshold (0.95), so I am uncertain whether additional erosion procedures (i.e., in addition to the default) are necessary. I am also uncertain how the default erosion (30mm) corresponds to masks twice or four times eroded, like mentioned in the Power paper referenced above.

I'd really appreciate your help!

Cheers,
Mihai

CPAC failed on example data

Xin Niu — Fri, 25 Feb 2022 22:08:08 GMT

Hi,

I am trying to run cpac with only one subject for testing purpose.

My data is organized in BIDS format as below:

ram/USERS/xin/BrainImaging_UNC/cpac_test/sub-0003107/func/*.nii
ram/USERS/xin/BrainImaging_UNC/cpac_test/sub-0003107/func/*.json
and
ram/USERS/xin/BrainImaging_UNC/cpac_test/sub-0003107/anat/*.nii

OSError: Didn't find any files in /ram/USERS/xin/BrainImaging_UNC/cpac_test. Please verify that the path is typed correctly, that you have read access to the directory, and that it is not empty.

I used hard link to copy the file to cpac_test. Is this the reason for not finding the files?
Thank you!

Best,
Xin

M-SDB

mk7kc — Fri, 26 Mar 2021 16:55:26 GMT

Hello.
I have a question as to what M-SDB stands for. My c-pac output directories get named in the following way: _selector_CSF-2mmE-M_aC-CSF+WM-2mm-DPC5_[b]M-SDB[/b]_P-2_BP-B0.01-T0.1

I understand which regressors each of the letters stand for except for M-SDB (in bold). I am guessing it is related to motion and the Friston 24-Parameter Model. I was hoping for confirmation that it means that motion-related artifacts were regressed out during the preprocessing and that I would not need to regress them out during the subsequent statistical analysis step.

My regressors from the pipeline config file are listed below:
- Bandpass:
bottom_frequency: 0.01
top_frequency: 0.1
CerebrospinalFluid:
erode_mask: true
extraction_resolution: 2
summary: Mean
Motion:
include_delayed: true
include_delayed_squared: true
include_squared: true
PolyOrt:
degree: 2
aCompCor:
extraction_resolution: 2
summary:
components: 5
method: DetrendPC
tissues:
- WhiteMatter
- CerebrospinalFluid

Thank you.

C-PAC slow runtime

Uri Elias — Sat, 21 Mar 2020 9:38:53 GMT

Hi all,

Before installing C-PAC,I've ran the old Fcon1000 preprocessing script on some FCon1000 data. Onmy machine (Win10+WSL), it took ~7 mins per scan.
Moving to C-PAC (installed via docker), running the default pipeline on the same data took ~1 HOUR per scan.

Does it make sense, have you experienced something similar?
Is there any explanation for the full order of magnitude change slow down?
Is there something I may not be doing correctly?

Thank you

Postdoctoral position: fMRI in OCD at Nathan Kline Institute (New York)

Emily Stern — Wed, 06 Jun 2018 20:52:45 GMT

Please see the below advertisement for a postdoctoral position studying the neural mechanisms of obsessive-compulsive disorder using fMRI at the Nathan Kline Institute!

[b]Post-doctoral Fellowship in Neuroimaging in Obsessive-Compulsive and Related Disorders[/b]
[b]Psychiatric NeuroCognition Lab[/b]
[b]The Nathan S. Kline Institute for Psychiatric Research (Orangeburg, NY)[/b]
[b]
[/b]
The Psychiatric NeuroCognition laboratory, under the direction of Dr. Emily Stern at the Nathan S. Kline Institute for Psychiatric Research, is seeking a post-doctoral fellow to work on projects investigating brain function and structure in obsessive-compulsive disorder (OCD) using task-based fMRI, resting-state functional connectivity, and diffusion imaging approaches. Broadly speaking, the lab's work uses a cognitive neuroscience approach to investigating brain function in psychiatric disorders, with a focus on using fMRI to tap into core constructs relevant for OCD and related disorders including interoception, rumination, decision-making, and reward processing. The person in this position will primarily work on two ongoing NIMH funded projects using fMRI to investigate neural mechanisms of sensory and other key symptoms in OCD, including a clinical trial in OCD and Tourette's Disorder testing the effects of a novel treatment on symptoms and brain function. In addition to working on these funded studies, there will also be opportunity for the fellow to develop new studies and/or analyses relevant to the lab's work and to obtain pilot research funding offered through the institute.

NKI is a research facility of the New York State Office of Mental Health, one of only two state-funded research institutes in the U.S. dedicated entirely to investigating biological mechanisms and treatments for psychiatric disorders. NKI was founded in 1952 by Dr. Nathan Kline, whose pioneering research into psychopharmacological management of schizophrenia and major depression revolutionized treatments for serious mental illness. The Psychiatric NeuroCognition lab is part of the Clinical Research Department at NKI, which contains several investigators, trainees, and staff working on studies examining treatments and biomarkers in a variety of disorders (depression, schizophrenia, autism spectrum disorder). Research facilities at NKI include a 3T Siemens MRI scanner for human neuroimaging, an MRI-compatible EEG system, and equipment for advanced neuromodulation techniques including transcranial magnetic stimulation and transcranial direct current stimulation. NKI also has an active animal research program with a 7T small animal scanner. For the past 30 years, NKI's academic affiliate has been the New York University School of Medicine (NYUSoM) located in midtown Manhattan, and investigators and trainees at NKI have access to many of NYUSoM's research tools and resources.

Approximately 20 miles north of Manhattan, NKI is located in suburban Rockland County (in Orangeburg, NY), home to many beautiful hiking and cycling trails overlooking the Hudson River. Although several towns in Rockland County are accessible to Manhattan via public transportation on New Jersey Transit, a driver's license and access to a car is highly recommended for anyone considering this position.

Interested candidates should meet the follow criteria:

[b]Required:[/b]
" Ph.D. in experimental psychology, clinical psychology, or cognitive neuroscience
" Moderate or better proficiency with functional MRI data analysis and presentation

[b]Other desirable qualifications:[/b]
" Strong interest in psychiatric disorders and applying cognitive neuroscientific techniques to OCD and related disorders
" Track record of publishing and presenting fMRI data
" Experience with 2 or more standard fMRI packages for data processing (SPM, FSL, AFNI, Conn, Freesurfer, DPABI/DPARSF, C-PAC) and/or experience with both task-based and resting-state functional connectivity methods
" Experience with task programing software such as EPrime, Presentation, or PsychoPy
" Good knowledge of basic statistical concepts and approaches
" Experience with (or desire to learn) basic programming in MATLAB, bash, and/or python
" Knowledge or experience in psychiatric/psychological assessment and diagnosis would be a plus but is not required

[b]Duties of the position include:[/b]
" Managing and analyzing task-activation and functional connectivity data, programming fMRI and behavioral experiments, running subjects through experimental protocols
" Reading the literature and publishing behavioral and neuroimaging findings in top tier journals
" Presenting results at conferences and seminars
" Developing new studies or analyses relevant for the ongoing work in the lab and applying for pilot and post-doctoral grants to support this work
" Assisting with grant applications, NIH progress reports, and IRB applications
" Assisting in the training of junior lab members including research coordinators and interns

Salary is competitive and commensurate with experience. Fellows would be employed by the Research Foundation for Mental Health (RFMH), which offers a full benefits package with medical and dental benefits. The position starts immediately and is for 1 year initially, with the opportunity to extend based on performance. Applications should be emailed to [b]fmripostdocnki@gmail.com[/b] ('fmri' and 'postdoc' and 'nki') and include a cover letter, CV, and the names and contact information for two referees.

RE: CPAC Benchmark Download?

Jane Zhang — Sat, 31 Dec 2016 16:44:53 GMT

Hi,

I had the same problem when I tried to download the CPAC benchmark data. The error message says, "Access Denied. Please contact your web hosting service provider for assistance." Could CPAC staff help me download the data?

Thanks, and Happy New Year!

Jane

Problems with portalocker.py

Christina Fales — Fri, 19 Feb 2016 19:32:58 GMT

Hi CPAC guys/gals:

A couple of months ago I asked about problems getting file locks when CPAC calls portalocker.py. You directed me to the Nipype folks who wrote the routine. I didn't get very far with them (and also have had to focus on other tasks for a while) so I am just now turning back to you. The problem is that portalocker.py uses the flock method when calling for locks, and flock apparently doesn't work with NFS. So I attempted to modify portalocker.py on our system so that it uses fcntl.lockf() instead. No luck. Still getting an Exception #37 ("No locks available"). (I've attached the modified version of portalocker.py... my changes are indicated with a "CF" comment.)

We also tried giving the various NFS servers the "nolock" attribute, which I gather should allow simultaneous locking of the log file, as along as the different users are on separate machines. (I'm not sure if this helps exactly, as the two accesses could very likely originate on the same machine.)

Could you offer any suggestions?

Thanks very much,
Christina Fales
Div. of Psychiatry Research
Feinstein Inst for Med Rsch
Zucker Hillside Hospital
265-31 76th Avenue
Glen Oaks, NY 11004

RE: CPAC Benchmark Download?

Chris Webb — Fri, 01 May 2015 15:39:14 GMT

Can I be granted access to the dataset to test my installation?

RE: Running cpac: paths/dependencies?

Naj Mahani — Fri, 07 Nov 2014 16:06:17 GMT

I think I found the problem;
chmod -R 777 ~/.matplotlib

But I found this by re-installing cpac using: ./cpac_install.sh

it's now working with just cpac_gui.

Running cpac: paths/dependencies?

Naj Mahani — Wed, 05 Nov 2014 20:51:51 GMT

Hi

I am hoping someone can tell me what it is I am doing wrong and how I can debug myself :)

I have tried to follow the installation instructions on the githu (http://fcp-indi.github.io/docs/user/install.html); and I run into the following problem:
[color=#800000][b]1) When running cpac_gui, I get: (Yes I have installed afni)[/b][/color]
Traceback (most recent call last):
File "/usr/local/bin/cpac_gui", line 3, in
import CPAC
File "/usr/local/lib/python2.7/dist-packages/CPAC/__init__.py", line 18, in
import anat_preproc, \
File "/usr/local/lib/python2.7/dist-packages/CPAC/anat_preproc/__init__.py", line 1, in
from anat_preproc import create_anat_preproc
File "/usr/local/lib/python2.7/dist-packages/CPAC/anat_preproc/anat_preproc.py", line 2, in
from nipype.interfaces.afni import preprocess
ImportError: No module named nipype.interfaces.afni
I suspect this has to do with incompatible paths, but I am not sure how to fix that actually.

[color=#800000]2) I wrote a py script (cpac.py) consisting of the following commands (pointing to the right path)[/color]
[color=#800000]import sys[/color]
[color=#800000]sys.path.insert(0,'/custom/path/to/isolated_nipype/site-packages')[/color]
[color=#800000]sys.path.insert(0,'/custom/path/to/isolated_CPAC/site-packages')[/color]
[color=#800000]import nipype[/color]
[color=#800000]import CPAC[/color]
[color=#800000]CPAC.GUI.run()[/color]

[color=#800000]this gets cpac_gui run; but[b] only through sudo!![/b][/color]

Without sudo I got the following:

Traceback (most recent call last):
File "cpac.py", line 6, in
import CPAC
File "/opt/cpac/lib/python2.7/site-packages/CPAC/__init__.py", line 18, in
import anat_preproc, \
File "/opt/cpac/lib/python2.7/site-packages/CPAC/qc/__init__.py", line 1, in
from .utils import *
File "/opt/cpac/lib/python2.7/site-packages/CPAC/qc/utils.py", line 3, in
import matplotlib
File "/usr/lib/pymodules/python2.7/matplotlib/__init__.py", line 774, in
rcParams = rc_params()
File "/usr/lib/pymodules/python2.7/matplotlib/__init__.py", line 692, in rc_params
fname = matplotlib_fname()
File "/usr/lib/pymodules/python2.7/matplotlib/__init__.py", line 604, in matplotlib_fname
fname = os.path.join(get_configdir(), 'matplotlibrc')
File "/usr/lib/pymodules/python2.7/matplotlib/__init__.py", line 253, in wrapper
ret = func(*args, **kwargs)
File "/usr/lib/pymodules/python2.7/matplotlib/__init__.py", line 475, in _get_configdir
raise RuntimeError("'%s' is not a writable dir; you must set %s/.matplotlib to be a writable dir. You can also set environment variable MPLCONFIGDIR to any writable directory where you want matplotlib data stored "% (h, h))
RuntimeError: '/home/nmahani' is not a writable dir; you must set /home/naj/.matplotlib to be a writable dir. You can also set environment variable MPLCONFIGDIR to any writable directory where you want matplotlib data stored

Much thanks