AHBAprocessing

This repository provides Matlab code for reproducing a range of analyses involving processing Allen Human Brain Atlas gene expression data and running a range of analyses to evaluate the effect of different processing choices. The code was verified using Matlab 2016b, 2017b, 2018a.

Please read our paper, and if you use this code, please cite our paper: A. Arnatkeviciute, B.D. Fulcher, A. Fornito. A practical guide to linking brain-wide gene expression and neuroimaging data.

NOTE: The code has been updated on the 28th August 2018 - in the previous version gene ordering in the ROI x gene matrices did not correspond to the gene information provided in the probeInformation structure. Now this issue is fixed. If you processed the data before this date, please re-process it using the updated code.

Contact Aurina Arnatkeviciute by email.

Dependencies:

Brain Connectivity Toolbox, Version 2017-15-01. Toolbox Fast Marching, Version 1.2.0.0

Data files

Data files required for this project are hosted on this figshare repository. Please download and unzip the data in the root directory.

Data processing

After retrieving data add relevant paths using AddPaths.m, data can be generated by running processingPipeline.m with selected options. In order to calculate distances on the cortical surface, toolbox_fast_marching toolbox needs to be installed by running the following lines within matlab from the code/peripheralFunctions/toolbox_fast_marching directory:

mex -setup
compile_mex;

This script will run 4 data processing functions:

S1_extractData - extracts the data from excel files and saves it in Matlab format
S2_probes - performs probe selection and data filtering if required
S3_samples2parcellation - assigns samples to the regions of interest in the parcellation
S4_normalisation - performs data normalisation and calculates correlated gene expression and also provides data in the region x gene format.

See README_processingPipeline.txt for more information.

To generate data for all four parcellations using selected options use processingPipeline_4parcellations.m.

Analysis

Figures in our paper can be regenerated using makeFigures.m in the plotting folder. All scripts should be run from the root directory. Other relevant analysis scripts are in dataProcessing.

Example usage:

After downloading data from figshare to the root directory, add the contents of the root directory to the path using AddPaths function.

To enable distance calculation on the cortical surface, install toolbox_fast_marching toolbox. To install the toolbox run the following lines within matlab from the code/peripheralFunctions/toolbox_fast_marching directory (for more information refer to documentation):

mex -setup
compile_mex;