Related Experiment Video
Updated: Jan 17, 2026
06:26
Meta-analysis of Voxel-Based Neuroimaging Studies using Seed-based d Mapping with Permutation of Subject Images SDM-PSI
Published on: November 27, 2019
77.2K
COMBINING META- AND MEGA- ANALYTIC APPROACHES FOR MULTI-SITE DIFFUSION IMAGING BASED GENETIC STUDIES: FROM THE
Neda Jahanshad1,2, Peter Kochunov3, Thomas E Nichols4,5
1Imaging Genetics Center, Institute of Neuroimaging Informatics, USC Keck School of Medicine, Los Angeles, CA, USA.
Proceedings. IEEE International Symposium on Biomedical Imaging
|September 18, 2025
Summary
Combining multi-site neuroimaging data using meta- and mega-analyses improves statistical power for heritability estimation. This approach enhances the precision of genetic findings in brain measures across diverse populations.
Area of Science:
- Neuroimaging Genetics
- Quantitative Trait Analysis
- Biostatistics
Background:
- Meta-analysis combines results from multiple studies without raw data access.
- Multi-site meta-analysis is vital for imaging genetics due to limited single-site sample sizes.
- Mega-analysis, sharing raw data, offers improved power and precision for global effect estimation.
Purpose of the Study:
- To estimate heritability of brain measures using fractional anisotropy (FA) maps.
- To compare the effectiveness of meta-analysis, mega-analysis, and a combined approach.
- To leverage multi-site data for robust genetic association studies in neuroimaging.
Main Methods:
- Utilized fractional anisotropy (FA) maps from 5 independent studies (N=2,203 subjects).
- Performed meta-analysis, mega-analysis, and a hybrid approach combining both methods.
- Investigated the statistical power and precision gains from different data combination strategies.
Main Results:
- A combination of mega- and meta-analytic approaches demonstrated potential for increased statistical power.
- Mega-analysis, when feasible with raw data sharing, enhances power and precision in estimating genetic effects.
- The study successfully estimated heritability of brain measures across a wide age range (9-85 years).
Conclusions:
- Hybrid meta- and mega-analytic strategies can optimize statistical power in neuroimaging genetics.
- Data sharing and collaborative analysis are crucial for advancing the field of imaging genetics.
- These methods provide a framework for more powerful and precise heritability estimation in large-scale neuroimaging studies.
