212

Bioinformatics of the Brain

Wang and coworkers performed NGS using whole blood samples from

ADHD patients and their healthy counterparts [93]. 13 miRNAs were dis-

covered as possible indicators for ADHD. In another study, Mccaffrey and

associates performed whole blood RNA sequencing on ADHD patients under

case-control [94]. They identified putative functions for a number of genes with

differential expression, including ABCB5, RGS2, GAK, and GIT1, which have

been linked mechanistically to molecular pathways associated with behavioral

control and ADHD in the past.

8.6

Integration of Brain Transcriptomics and Imaging

Data

There comes a time where data from only one source is not enough to un-

derstand the big picture anymore and integration of several types of data is

necessary to move forward. Integrating brain transcriptomics with neuroimag-

ing data advanced with the public unveiling of the Allen Human Brain Atlas

(AHBA) dataset in 2012. This dataset contained histology data, structural

MRI (sMRI), and whole-brain microarray transcriptome data collected from

healthy mature human subjects [95]. Since then, attempts were made to inte-

grate imaging with transcriptomics to shed more light into the etiology and

progression of these brain diseases and disorders as well as diagnostic and ther-

apeutic studies. For instance, Adewale and coworkers proposed a spatiotem-

poral brain model that takes into consideration the direct interaction between

numerous RNA transcripts and macroscale imaging techniques like MRI and

PET [96]. In another study, Wu and coworkers worked on a so-called federated

model in detection of genomic and transcriptomic factors associated with AD

using sMRI, GWAS (genome-wide association studies), and transcriptomics

data [97]. It is important to interpret the interplay of biological factors at var-

ious spatial resolutions. This area is also called imaging genetics, which is the

application of neuroimaging tools to examine how genetic differences affect

brain structure or function in order to gain an insight into how these varia-

tions affect behavior and disease phenotypes [98]. Research in this field has

been gaining momentum in recent years [99–107]. There are also online tools

specialized in this area. One example of such tools is the Neuroimaging Infor-

matics Tools and Resources Clearinghouse (www.nitrc.org), often known as

NITRC-R. It is a collection of resources for neuroimaging, including data sets,

software for analysis, and computer power. The research focus of NITRC com-

prises software tools, data, and computational resources for MR, PET/SPECT

(Single-photon emission computed tomography), CT (computerized tomogra-

phy), EEG (electroencephalogram)/MEG (Magnetoencephalography), optical

imaging, clinical neuroimaging, computational neuroscience, and imaging ge-

nomics.