A FreeSurfer view of the cortical transcriptome generated from the Allen Human Brain Atlas

The Allen Human Brain Atlas provides an anatomically comprehensive view of gene expression in the brain (Hawrylycz et al., 2012). The complete transcriptome dataset consists of 58,692 measurements of gene expression in 3702 brain samples obtained from 6 individuals. The resulting product, over 200 million gene expression values, can be overwhelming for neuroscientists seeking to use the data. For example, data for a single gene consist of three interlinked files when downloaded from the Allen Institute website. The local expression values are associated with X, Y, and Z coordinates in the MNI152 atlas (Mazziotta et al., 2001), thus allowing one to map these values to other brain atlases. To reduce the complexity of the Allen data, we have summarized the data into the Desikan– Killiany cortical atlas built into the FreeSurfer software for automatic labeling of regions of interest (Desikan et al., 2006). FreeSurfer allows one to segment magnetic resonance images (MRIs) into 68 cortical regions and to estimate their cortical thickness, surface area and volume (http://surfer.nmr. mgh.harvard.edu/). These cortical measurements are stable and agree with past histological studies (Fischl and Dale, 2000; Han et al., 2006; Scholtens et al., 2015). FreeSurfer has been used widely, with over 600 reports covering a broad range of topics that include brain development, aging and a variety of brain disorders (PubMed, May 2015). Adding the perspective of gene expression could facilitate interpretation of these reports. For example, we recently found that the magnitude of group differences in cortical thickness between cannabis users and non-users correlates with regional variations in expression levels of cannabis receptor 1 (CNR1) (French et al., 2015).