Introduction and Setup

Introduction

This is a tutorial that illustrates the steps involved in taking raw data from an fMRI study and running basic preprocessing to prepare fMRI data for basic GLM analysis. It includes:

• conversion from DICOMs to NIfTIs
• basic preprocessing of functional data, including motion correction and quality assurance
• basic preprocessing of anatomical data, including FreeSurfer
• unwarping and registration of functional data

For an overview of the steps, see the conceptual overview slides.

The tutorial currently does not cover the following, but could be expanded in the future to cover:

• creation of an anatomical template using ANTS
• univariate analysis using FEAT
• multivariate analysis using pymvpa

Installation

First, follow the installation instructions to:

• set up a TACC account
• get access to Lonestar 6
• use TACC's module system to load dependencies
• set up a Python 3 virtual environment with fPrep

Tutorial

The tutorial is divided into six parts. There is a copy of the files taken at the end of each part on corral; for example, the results of Part 2 are in: /corral-repl/utexas/prestonlab/preproc/part2. These are just snapshots for people to look at (for example, in fslview); do not modify these files. All commands to process the sample data should be run on a copy of the data in your WORK directory, not on corral.

You can access the steps you need quickly here:

Part 1: Getting Data Part 2: Converting to NIfTI Part 3: Organizing Data Part 4: Basinc BOLD Preprocessing Part 5: Anatomical Preprocessing Part 6: BOLD Unwarping and Registration