Multivariate Analysis and Modelling of multiple Brain endOphenotypes: Let's MAMBO!

Abstract

Imaging genetic studies aim to test how genetic information influences brain structure and function by combining neuroimaging-based brain features and genetic data from the same individual. Most studies focus on individual correlation and association tests between genetic variants and a single measurement of the brain. Despite the great success of univariate approaches, given the capacity of neuroimaging methods to provide a multiplicity of cerebral phenotypes, the development and application of multivariate methods become crucial. In this article, we review novel methods and strategies focused on the analysis of multiple phenotypes and genetic data. We also discuss relevant aspects of multi-trait modelling in the context of neuroimaging data.

Keywords: Genetics; Image-derived phenotype; Imaging genetics; Multiple phenotypes; Multivariate modelling; Neuroimaging.

Fig. 1

Framework of analytical strategies for multiple phenotype assessment. *Year of these publications corresponds to a literature review describing this family of methods. Legend: BGSMTR: Bayesian group sparse multi-task regression model; CCA: Canonical Correlation Analysis; GLM: General Linear Model; ICA: Independent Component Analysis; ICA-MFA: Independent Multiple Factor Association Analysis for Multiblock Data; L2R2: Bayesian longitudinal low-rank regression; LMM: Linear Mixed Model; MANOVA: Multivariate Analysis of Variance; MFA; Multiple Factor Analysis; PCA: Principal Component Analysis; sRRR: sparse Reduced Rank Regression. This schematic representation was created with Biorender (©BioRender - biorender.com).