Variability in Resting State Network and Functional Network Connectivity Associated With Schizophrenia Genetic Risk: A Pilot Study

Jiayu Chen, Barnaly Rashid, Qingbao Yu, Jingyu Liu, Dongdong Lin, Yuhui Du, Jing Sui, Vince D. CalhounFrontiers in Neuroscience 2018 · DOI

Introduction: The Dimensionality Challenge in Imaging Genetics

Imaging genetics is a powerful field that seeks to map the influence of genetic variants on brain structure and function. This approach is particularly vital for understanding psychiatric disorders like schizophrenia (SZ), which are known to have a strong polygenic basis. However, researchers face a significant technical hurdle: the "dimensionality mismatch." Genetic data is typically structured in two dimensions (Subject × Genetic Variant), whereas functional MRI (fMRI) data is four-dimensional (Subject × Voxel × Time).

To bridge this gap, scientists must derive meaningful "summary features" from fMRI data that can be compared against genetic markers at the subject level. While traditional studies often focus on the average strength of connectivity, these measures can overlook the subtle, individualized deviations in brain organization that characterize complex disorders.

Methodology: Quantifying Brain Network Variability

In this pilot study, researchers proposed a novel summary feature: variability in resting-state networks (RSNs) and functional network connectivity (FNC). Rather than measuring the absolute strength of a network, this approach quantifies how much an individual's functional brain architecture deviates from a cohort-common "blueprint."

The proposed workflow for estimating similarity matrices and network variability through Group ICA and back reconstruction.

The study utilized a Group Independent Component Analysis (GICA) framework on a dataset of 305 subjects (171 healthy controls and 134 patients with schizophrenia). The process involved:

  1. Network Extraction: Identifying 75 RSNs representing functional units of the brain.
  2. Template Creation: Establishing a group-level aggregate ($S_{agg}$) for both RSNs and FNC.
  3. Variability Calculation: Measuring the "distance" between each subject's individual map ($S_i$) and the group template using metrics like Euclidean distance and Pearson correlation.

Key Findings: Deviation as a Biomarker

The study revealed that "straying from the norm" is a significant biological signal. Patients with schizophrenia exhibited consistently higher variability across multiple brain networks compared to healthy controls.

Comparison of network variation between healthy controls (HC) and schizophrenia patients (SZ) across ten significant independent components, showing consistently higher variability in the patient group.

Increased Network Variation in Schizophrenia

As shown in Figure 2, ten specific networks—predominantly those involved in visual and sensorimotor processing—showed significantly higher variation in the SZ group ($p$-values ranging from $10^{-4}$ to $10^{-8}$). This increased deviation suggests that the functional organization of the brain in schizophrenia is less "standardized" and more idiosyncratic than in healthy individuals.

The Cost of Variability: Neural Efficiency

The researchers found that this variability is not merely random noise; it has functional consequences. Higher deviation from the group-common pattern was significantly correlated with lower Global Efficiency. Global efficiency is a measure of how effectively the brain integrates information across distant regions. This suggests that the unconventional network organization seen in SZ patients may come at the cost of efficient neural communication.

Genetic Underpinnings

Crucially, the study linked these imaging features to genetic risk. Variability in the parahippocampal gyrus was associated with the SNP rs11926768 in the TRANK1 gene, a known risk factor for both bipolar disorder and schizophrenia. Furthermore, a higher Polygenic Risk Score (PRS)—a cumulative measure of genetic burden for SZ—was associated with increased FNC variability. This provides preliminary evidence that the brain's functional "unconventionality" is biologically regulated and tied to genetic susceptibility.

Conclusion and Future Directions

This pilot study demonstrates that variability in RSN and FNC is a promising biomarker for schizophrenia. By capturing how individual brains deviate from a common template, researchers can better bridge the gap between high-dimensional imaging data and genetic risk factors.

While the results are compelling, the authors note that future work should utilize larger, "gold standard" healthy databases to refine the reference templates. Additionally, exploring "dynamic" variability—how these networks fluctuate over the course of a scan—could provide even deeper insights into the functional instability associated with psychiatric risk.

Summary updated

Abstract

Imaging genetics posits a valuable strategy for elucidating genetic influences on brain abnormalities in psychiatric disorders. However, association analysis between 2D genetic data (subject × genetic variable) and 3D first-level functional magnetic resonance imaging (fMRI) data (subject × voxel × time) has been challenging given the asymmetry in data dimension. A summary feature needs to be derived for the imaging modality to compute inter-modality association at subject level. In this work, we propose to use variability in resting state networks (RSNs) and functional network connectivity (FNC) as potential features for purpose of association analysis. We conducted a pilot study to investigate the proposed features in a dataset of 171 healthy controls and 134 patients with schizophrenia (SZ). We computed variability in RSN and FNC in a group independent component analysis framework and tested three types of variability metrics, namely Euclidean distance, Pearson correlation and Kullback-Leibler (KL) divergence. Euclidean distance and Pearson correlation metrics more effectively discriminated controls from patients than KL divergence. The group differences observed with variability in RSN and FNC were highly consistent, indicating patients presenting increased deviation from the cohort-common pattern of RSN and FNC than controls. The variability in RSN and FNC showed significant associations with network global efficiency, the more the deviation, the lower the efficiency. Furthermore, the RSN and FNC variability were found to associate with individual SZ risk SNPs as well as cumulative polygenic risk score for SZ. Collectively the current findings provide preliminary evidence for variability in RSN and FNC being promising imaging features that may find applications as biomarkers and in imaging genetic association analysis.
BibTeX
@inproceedings{chen2018variability,
  author = {Jiayu Chen and Barnaly Rashid and Qingbao Yu and Jingyu Liu and Dongdong Lin and Yuhui Du and Jing Sui and Vince D. Calhoun},
  title = {{Variability in Resting State Network and Functional Network Connectivity Associated With Schizophrenia Genetic Risk: A Pilot Study}},
  booktitle = {Frontiers in Neuroscience},
  year = {2018},
  doi = {10.3389/fnins.2018.00114},
}
Figure