Whole genome association study of brain‐wide imaging phenotypes: A study of the ping cohort

Abstract

Neuropsychological disorders have a biological basis rooted in brain function, and neuroimaging data are expected to better illuminate the complex genetic basis of neuropsychological disorders. Because they are biological measures, neuroimaging data avoid biases arising from clinical diagnostic criteria that are subject to human understanding and interpretation. A challenge with analyzing neuroimaging data is their high dimensionality and complex spatial relationships. To tackle this challenge, we introduced a novel distance covariance tests that can assess the association between genetic markers and multivariate diffusion tensor imaging measurements, and analyzed a genome-wide association study (GWAS) dataset collected by the Pediatric Imaging, Neurocognition, and Genetics (PING) study. We also considered existing approaches as comparisons. Our results showed that, after correcting for multiplicity, distance covariance tests of the multivariate phenotype yield significantly greater power at detecting genetic markers affecting brain structure than standard mass univariate GWAS of individual neuroimaging biomarkers. Our results underscore the usefulness of utilizing the distance covariance to incorporate neuroimaging data in GWAS.

TransferGWAS: GWAS of images using deep transfer learning

“…Such quantities have been obtained manually (e.g., segmenting organ volumes using graphics software), by automated software tools 1 , or deep learning models trained to predict segmentation masks or other biomarkers. [2][3][4][5][6][7] However, such methods are limited in the scope of questions they can address and prohibit the detection of associations between unspecified variation of the studied organ and the genetic variation. Further challenges are that (i) deep learning methods require large amounts of labeled data, which can be very costly to produce, (ii) the extracted sets of biomarkers are generally study-and disease-specific 8,9 , and (iii) deep learning-based feature extraction often leads to noisier estimates of the trait under investigation compared to hand-labeled and quality-controlled labels.…”

transferGWAS: GWAS of images using deep transfer learning

“…Current approaches to GWAS of imaging data have been largely focused on extracting fixed quantities of interest from the images, such as organ volumes, distances, and sizes. Such quantities have been obtained manually (e.g., segmenting organ volumes using graphics software), by automated software tools [16], or deep learning models trained to predict segmentation masks or other biomarkers [21,15,35,41,3,17]. However, such methods are limited in the scope of questions they can address and prohibit the detection of associations between unspecified variation of the studied organ and the genetic variation.…”