An ICA with reference approach in identification of genetic variation and associated brain networks

Jingyu Liu, Mohammad M. Ghassemi, Andrew M. Michael, David Boutte, William M. Wells, Nora I. Perrone‐Bizzozero, Fabìo Macciardi, Daniel H. Mathalon, Judith M. Ford, Steven G. Potkin, Jessica A. Turner, Vince D. CalhounFrontiers in Human Neuroscience 2012 · DOI

Abstract

To address the statistical challenges associated with genome-wide association studies (GWAS), this paper presents an independent component analysis (ICA) with reference approach to target specific genetic variations and associated brain networks. By using a small set of empirically chosen single nucleotide polymorphisms (SNPs) as a reference, the method extracts genetic components that maximally represent characteristics of interest. This approach was evaluated on simulated data and a real dataset of 48 schizophrenia patients and 40 healthy controls. Results identified a genetic component differentiating patients from controls ($p < 4 \times 10^{-17}$) and a significantly associated functional brain network ($p < 1 \times 10^{-4}$) located in the thalamus and cingulate gyri.

Introduction

Bridging the gap between massive genomic datasets and functional brain imaging remains a significant hurdle in neuroscience. Genome-wide association studies (GWAS) often struggle with statistical power due to the sheer volume of variables (millions of SNPs) relative to subject numbers, causing subtle signals to be lost in noise. While standard Independent Component Analysis (ICA) can find patterns, it is a "blind" technique that prioritizes factors explaining the most variance, potentially overlooking disease-related factors with low total genomic variance.

To solve this, researchers developed "ICA with reference," a semi-blind source separation technique. This method uses a small set of "hint" SNPs to guide the extraction of meaningful genetic factors, allowing for a more targeted exploration of the genome's effect on the brain.

Methodology

The core innovation of this study is the constrained ICA algorithm. The process involves three main stages:

  1. Reference Generation: A reference signal is created based on a subset of SNPs showing initial differences between groups (e.g., via a t-test).
  2. Guided Extraction: The standard FastICA algorithm is modified with a constraint that penalizes the extracted component if it deviates significantly from the reference signal.
  3. Multimodal Linking: The loading coefficients of the extracted genetic component are correlated with the loading coefficients of brain networks derived from fMRI data using standard Infomax ICA.
The flowchart of the proposed method.

Results and Discussion

The method was validated using both simulated and real-world data from the FBIRN Phase II study, involving 88 subjects (48 Schizophrenia patients, 40 Healthy Controls).

Simulation and Sensitivity

Simulation demonstrated that ICA with reference can successfully extract "quiet" genetic factors that standard ICA misses. Crucially, the algorithm remained robust even when the reference signal was only 25% accurate, proving its utility in finding a "needle in a haystack."

Schizophrenia Application

In the real-world dataset (~300,000 SNPs and fMRI data from an auditory oddball task), the approach yielded several key findings:

  • Genetic Differentiation: The extracted genetic component differentiated patients from controls with extreme significance ($p < 4 \times 10^{-17}$).
  • Brain-Gene Association: This genetic factor correlated significantly ($r = 0.39, p < 1 \times 10^{-4}$) with a functional brain network involving the thalamus and the anterior/posterior cingulate gyri.
  • Identified Gene Clusters: Top-contributing SNPs clustered in two regions: 7q21 (including CDK14) and 5q35 (including GRM6, a glutamate receptor gene).

Conclusion

ICA with reference provides a powerful tool for identifying polygenic clusters and their functional manifestations in the brain. While limited by sample size, the study successfully identified a biological bridge between DNA and clinical symptoms in schizophrenia. Future applications could utilize references derived from known biological pathways (e.g., dopamine or glutamate) to enable more hypothesis-driven genomic exploration.

Summary updated

Abstract

To address the statistical challenges associated with genome-wide association studies, we present an independent component analysis (ICA) with reference approach to target a specific genetic variation and associated brain networks. First, a small set of single nucleotide polymorphisms (SNPs) are empirically chosen to reflect a feature of interest and these SNPs are used as a reference when applying ICA to a full genomic SNP array. After extracting the genetic component maximally representing the characteristics of the reference, we test its association with brain networks in functional magnetic resonance imaging (fMRI) data. The method was evaluated on both real and simulated datasets. Simulation demonstrates that ICA with reference can extract a specific genetic factor, even when the variance accounted for by such a factor is so small that a regular ICA fails. Our real data application from 48 schizophrenia patients (SZs) and 40 healthy controls (HCs) include 300K SNPs and fMRI images in an auditory oddball task. Using SNPs with allelic frequency difference in two groups as a reference, we extracted a genetic component that maximally differentiates patients from controls (p < 4 × 10(-17)), and discovered a brain functional network that was significantly associated with this genetic component (p < 1 × 10(-4)). The regions in the functional network mainly locate in the thalamus, anterior and posterior cingulate gyri. The contributing SNPs in the genetic factor mainly fall into two clusters centered at chromosome 7q21 and chromosome 5q35. The findings from the schizophrenia application are in concordance with previous knowledge about brain regions and gene function. All together, the results suggest that the ICA with reference can be particularly useful to explore the whole genome to find a specific factor of interest and further study its effect on brain.
BibTeX
@inproceedings{liu2012ica,
  author = {Jingyu Liu and Mohammad M. Ghassemi and Andrew M. Michael and David Boutte and William M. Wells and Nora I. Perrone‐Bizzozero and Fabìo Macciardi and Daniel H. Mathalon and Judith M. Ford and Steven G. Potkin and Jessica A. Turner and Vince D. Calhoun},
  title = {{An ICA with reference approach in identification of genetic variation and associated brain networks}},
  booktitle = {Frontiers in Human Neuroscience},
  year = {2012},
  doi = {10.3389/fnhum.2012.00021},
}
Figure