A novel classification framework for genome-wide association study of whole brain MRI images using deep learning

Shaojun Yu, Junjie Wu, Yumeng Shao, Deqiang Qiu, Zhaohui S. Qin

Abstract: Author summary Genome-wide association study (GWAS) is a powerful method to identify associations between genetic variants and traits such as height, weight and disease status. When applying to Magnetic resonance imaging (MRI) data, traditional GWAS methods often rely on simplified summaries of brain imaging data, potentially missing subtle but significant global patterns. We proposed and implemented a different strategy: training a machine learning model to distinguish MR images based on genetic variants. If MRI images labeled by the mutation status of a variant can be reliably distinguished using machine learning, we then hypothesized that this variant is likely to be associated with brain anatomy or function which is manifested in MRI brain images. By applying this method to data collected from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), we found new genetic variants highly likely to affect brain phenotypes. This innovative approach not only handles high-dimensional imaging data more effectively but also captures complex, non-linear relationships between genetic variants and various brain traits, offering a fresh perspective on neuroimaging genetics.

External IDs:dblp:journals/ploscb/YuWSQQ24