Abstract—Structural equation modeling (SEM) is a statistical technique widely used in quantitative genetics to measure genetic and environmental variances of human traits. Using SEM, the proportions of genetic and environmental influences can be separated from the phenotypic variance. However, the SEM softwares like Mx or LISREL were not designed for a big data analysis. They can hardly be applied for brain images that comprise hundreds of thousands of voxels. Here, to introduce SEM in the field of neuroimaging, we developed a simple code in MATLAB for multiple computations of Mx. Our method could estimate genetic and environmental variances of neural activations at 153,594 voxels of the whole brain, to be converted to brain images.

Index Terms—fMRI, genetics, structural equation modeling, twin.

Y. Y. Choi, J. I. Song, J. S. Chun, and W. K. Song are with BioImaging Research Center, Gwangju Institute of Science and Technology, Gwangju 500-712 Korea (e-mail: yuyongchoi@gist.ac.kr, jisong@gist.ac.kr, jschun@gist.ac.kr, wksong@gist.ac.kr).
K. H. Lee is with Dementia Center, Chosun University, Gwangju 501-759 Korea (e-mail: leekho@chosun.ac.kr).

 

Cite:Yu Yong Choi, Jong-In Song, Jang Soo Chun, Kun Ho Lee, and Woo Keun Song, "A Structural Equation Modeling Approach for the Estimation of Genetic and Environmental Effects from Twin fMRI Data," International Journal of Bioscience, Biochemistry and Bioinformatics vol. 3, no. 3, pp. 167-169, 2013.