lab.info: These are the utility functions for the data tracker.

In DecisionNeurosciencePsychopathology/DataTracker: DataTracker

Description

—————————————————————————–

SUMMARY:

—————————————————————————–

All functions for the Decision Neuroscience and Psychopathology Lab at the University of Pittsburgh Medical Center. This is a single-file R package that was developed to connect to research data data across protocols and modalities and even cloud storage resources by wrapping Rclone.

—————————————————————————–

GOAL:

—————————————————————————–

The design was meant to be simple, hence the single file. Abstract functions are to be built that can then be utilized to quickly access data in a uniform manner across protocols. This should, hopefully, reduce ad-hoc scripting and make data management easier, faster, more precise and, most importantly, extensible to new tasks, protocols, and data storage. Essentially, this single page should be the main resource to use and refer to for data management at the DNPL.

The only other files a user should need are:

1. A configuration json file This should be provided by the current maintainer(s) in a private GitHub repo. A private GitHub repo was chosen as it provides easy access for lab members by using git clone.

2. An rclone config file This is actually managed by rclone using the "rclone config" command. https://rclone.org/commands/rclone_config/

—————————————————————————–

MAINTAINERS:

—————————————————————————–

Current Maintainers:

Shane Buckley: tshanebuckley<at>gmail.com

Previous Maintainers:

NA

—————————————————————————–

FUNCTION NAMING RULES:

TODO: document all rules for function naming here.

—————————————————————————–

KEYWORDS and CONCEPTS: TODO: have someone else update tasks and protocols

—————————————————————————–

PROTOCOLS: masterdemo:

bsocial

ksocial

explore

explore2

momentum (not implemented yet)

pandea (not implemented yet)

TASKS: clock

trust

spott

CLINICAL DATA/PROTOCOL STATUS SOURCES: redcap Database for storing participant data by identifiers. Used to track participant participation in protocols along with demographic data (Master Demographic).

site: https://www.project-redcap.org/

documentation: https://wiki.uiowa.edu/display/REDCapDocs/REDCap

ctsi: https://ctsi.pitt.edu/guides-tools/data-management-resources/

pitt: https://www.ctsiredcap.pitt.edu/redcap/

R SDK for data fetching: https://ouhscbbmc.github.io/REDCapR/

Python SDK for fetching data: https://pycap.readthedocs.io/en/latest/

RAW IMAGING DATA SOURCES/FORMATS: meson Database used before xnat to get imaging data from the MRRC.

xnat Current database used to get imaging data from the MRRC.

Python SDKs: xnatpy: https://xnat.readthedocs.io/en/latest/

pyxnat: https://pyxnat.github.io/pyxnat/

MRRC XNAT: Script used to pull data from MRRC (DNPL fork): https://github.com/DecisionNeurosciencePsychopathology/dax/blob/main/bin/Xnat_tools/Xnatdownload

To Install: pip install git+https://github.com/DecisionNeurosciencePsychopathology/dax.git

Example usage:

Xnatdownload -p WPC-7341 -d /volume1/bierka_root/datamesh/RAW –subj all –sess all -s all -a all –rs all –ra all (Run as a cron job, the above function downloads all xnat data for bsocial.)

BRAIN IMAGING DATA STRUCTURE: bids Standardized format for brain imaging data.

BIDS main page: https://bids.neuroimaging.io/ BIDS documentation: https://bids-specification.readthedocs.io/en/stable/ Python SDK: documentation: https://github.com/bids-standard/pybids tutorial: https://notebook.community/INCF/pybids/examples/pybids_tutorial Conversion(heudiconv): documentation: https://heudiconv.readthedocs.io/en/latest/ tutorial: https://reproducibility.stanford.edu/bids-tutorial-series-part-2a/

IMAGING DATA PRE-PROCESSING: fmriprep Standardized preprocessing software that take BIDS-converted data as input.

fMRIPrep Documentation: https://fmriprep.org/en/stable/

clpipe: Wrapper to run fmriprep via a singularity container on a slurm cluster.

clpipe documentation: https://clpipe.readthedocs.io/en/latest/

clpipe github: https://github.com/cohenlabUNC/clpipe

DNPL fork: https://github.com/DecisionNeurosciencePsychopathology/clpipe

crc: University of Pittsburgh's HPC cluster for running slurm jobs.

crc documentation: https://crc.pitt.edu/ (Refer to this for Globus, Slurm, and general HPC usage documentation.)

psychiatry visualization portal: https://crc.pitt.edu/psych (GPU-accelerated desktop interface for image visualization via FSL and AFNI.)

longleaf: University of North Carolina Chapel Hill's HPC cluster for running slurm jobs.

LOCAL STORAGE: bierka 80TB Synology NAS backed up via tape by the OAC and to the DNPL DataTeam SharePoint. This is our long-term storage server.

milka 40TB Synology NAS used for local scratch space for the DNPL. This is our short-term storage server. (Currently backup of Bek)

bek Pegasus NAS containing legacy data.

skinner Main SharePoint site for DNPL collaboration and behavioral data upload.

rclone Software that allows the syncing and mounting of cloud storage.

documentation: https://rclone.org/

minio S3 compliant data bucket software (allows mounting Bierka via rclone).

documentation: https://docs.min.io/docs/minio-quickstart-guide.html

docker Containerization software (installed on Bierka).

documentation: https://docs.docker.com/

sharepoint Pitt's chosen cloud storage provided as of 2021.

—————————————————————————–

Usage

lab.info()

Examples

?lab.cfg

DecisionNeurosciencePsychopathology/DataTracker documentation built on Jan. 20, 2022, 4:51 p.m.