Variability in Resting State Network and Functional Network Connectivity Associated With Schizophrenia Genetic Risk:...

Area of Science:

• Neuroimaging
• Psychiatric Genetics
• Computational Neuroscience

Background:

• Imaging genetics aims to link genetic data with brain abnormalities in psychiatric disorders.
• Analyzing 2D genetic data and 3D functional magnetic resonance imaging (fMRI) data is challenging due to dimensional asymmetry.
• Summary features are needed for imaging data to enable subject-level association analysis.

Purpose of the Study:

• To propose and evaluate variability in resting state networks (RSNs) and functional network connectivity (FNC) as imaging features for association analysis.
• To investigate these features in a pilot study involving healthy controls and patients with schizophrenia (SZ).
• To explore the association of these features with genetic risk factors and network efficiency.

Main Methods:

• Utilized a group independent component analysis (ICA) framework to compute RSN and FNC variability.
• Tested Euclidean distance, Pearson correlation, and Kullback-Leibler (KL) divergence as variability metrics.
• Analyzed data from 171 healthy controls and 134 schizophrenia patients.

Main Results:

• Euclidean distance and Pearson correlation metrics effectively discriminated controls from patients.
• Patients showed greater deviation in RSN and FNC patterns compared to controls.
• RSN and FNC variability correlated with network global efficiency and schizophrenia (SZ) risk single nucleotide polymorphisms (SNPs) and polygenic risk scores.

Conclusions:

• Variability in RSN and FNC are promising imaging features for association studies in psychiatric disorders.
• These features demonstrate potential as biomarkers for schizophrenia.
• The findings support the utility of RSN and FNC variability in imaging genetics research.