egarza / sudmex_tms_main

Datasets and code for the preprint "Clinical and functional connectivity outcomes of 5-Hz repeated transcranial magnetic stimulation as an add-on treatment in cocaine use disorder: a double-blind randomized controlled trial."

SUDMEX_TMS: The Mexican dataset of an rTMS clinical trial on cocaine use disorder patients

Datasets and code for the preprint "Clinical and functional connectivity outcomes of 5-Hz repeated transcranial magnetic stimulation as an add-on treatment in cocaine use disorder: a double-blind randomized controlled trial".

Please let me know if something in the code is incomplete or files are missing or you have any problems.

About

The SUDMEX_TMS dataset is the result of a longitudinal clinical trial of cocaine use disorder patients that were treated with rTMS at 5-Hz on the left dorsolateral prefrontal cortex (lDLPFC) for a 2 week double-blind acute phase and an open-label maintenance phase that included clinical, cognitive and MRI data acquisition. The design was a double-blind placebo-controlled randomized controlled trial with parallel groups (acute phase). The study was done at the National Institute of Psychiatry in Mexico City by PIs Dr. Eduardo A. Garza-Villarreal and Dr. Ruth Alcala-Lozano. Each patient had more than one clinical and MRI session or time point from baseline (T0), 2 weeks (T1), 3 months (T2), 6 months (T3) and some patients had 12 months (T4). The T14 time point was only for patients in the Sham group who decided to continue the clinical trial with open-label rTMS. The T14 refers to 2 weeks after T1 (4 weeks after T0). The MRI sequences were: 1) T1-weighted, 2) rsfMRI 10 min, 3) HARDI-DWI multishell (next release).

For more information, check out the preprint.

Data

BIDS compliant raw MRI data are available at OpenNeuro.

Clinical data are in the folders data/clinical. The files are in CSV and labels are ran in the code/Rcode.Rmdfile.

The folder data/normative stores the normative lDLPFC seed cone and the resulting normative connectivity.

Code

The code is stored in code (easy). Clinical analysis can be run using the code/Rcode.Rmdfile by changing the root folder in the code. The lDLPFC and vmPFC correlations can be calculated using correlation_dlpfc.sh and correlation_vmpfc.sh. For that you need the derivatives from fmriprep and xcpEngine, which can be sent to your server (320 Gb). We also included the file acute_interaction.shwhich includes the Interaction analysis made in FSL step by step. Preprocessing was done using fmriprep, and the pipeline is the file fmriprepTMS.SGE. Processing was done using xcpEngine, and the pipeline is the file xcpEngineFuncTMS.SGE, while the xcp_Engine folder includes the cohort and designsfiles.

The tms_location folder has all the individual TMS locations in csv, the individual spheres and the code to produce them, as well as the images needed for our manuscript Figures. The script SimulateTargetTMS.m was used to create the cortical locations of the TMS localization in Matlab

Analysis

The folder includes the Acute Interaction analysis, with the design and contrast files used in FSL, the Regions of Interest and the resulting corrected and uncorrected maps. The Rmaps, Zmaps and Difference maps needed for the analysis can be found on Zenodo.

Derivatives

MRIQC derivatives can be found on Zenodo. fmriprep and xcpEngine derivatives are 320 Gb and couldn't be stored freely. Therefore, they are available on demand.

Links

1. Preprint: https://www.medrxiv.org/content/10.1101/2020.07.15.20154708v1
2. Raw data: https://openneuro.org/datasets/ds003037
3. Some derivatives and files: https://zenodo.org/record/3967527#.XzMDdXX0k5k
4. Brain maps in Neurovault: https://neurovault.org/collections/8519/