i# project-iota STAT 159/259: Reproducible and Collaborative Statistical Data Science

Fall 2015 UC Berkeley

This repository contains reproducible analysis based on dataset used in [Working Memory in Healthy and Schizophrenic Individuals] (https://openfmri.org/dataset/ds000115).

Installation

1. Clone our project repository: https://github.com/berkeley-stat159/project-iota
2. Install python modules with pip: pip install -r requirements.txt

General steps:

Download Dataset

• make dataset: Download the 4.2GB ds115_sub001-005.tgz and nipy.bic.berkeley.edu, then automatically unzip sub001, the only subject data that we worked with.

Generate Convolution

• make convo: Convolve study conditions with hemodynamic function(HDF). Included are the regular block design convolutions as well as a higher resolution convolutions for event-related convolutions.

Modeling & Analysis

• make modeling: Run all the regressions and related plots mentioned in our report.

Hypothesis Testing

• make testing: Run all the hypothesis tests and validations of our mode.

Report

• make report: Compile our final report with analysis results.

Thanks to Jarrod Millman, Matthew Brett, [Ross Barnowski] (https://github.com/rossbar) and [J-B Poline] (https://github.com/jbpoline) for their instructions throughout the semester.

Contributors

• Jie Li (Jay4869)
• Zeyu Li (lizeyuyuz)
• Qingyuan Zhang (amandazhang)
• Yun Chuan (ay2456)