Longitudinal Genotype-Phenotype Association Study via Temporal Structure Auto-learning Predictive Model

Abstract

With rapid progress in high-throughput genotyping and neuroimaging, imaging genetics has gained significant attention in the research of complex brain disorders, such as Alzheimer’s Disease (AD). The genotype-phenotype association study using imaging genetic data has the potential to reveal genetic basis and biological mechanism of brain structure and function. AD is a progressive neurodegenerative disease, thus, it is crucial to look into the relations between SNPs and longitudinal variations of neuroimaging phenotypes. Although some machine learning models were newly presented to capture the longitudinal patterns in genotype-phenotype association study, most of them required fixed longitudinal structures of prediction tasks and could not automatically learn the interrelations among longitudinal prediction tasks. To address this challenge, we proposed a novel temporal structure auto-learning model to automatically uncover longitudinal genotype-phenotype interrelations and utilized such interrelated structures to enhance phenotype prediction in the meantime. We conducted longitudinal phenotype prediction experiments on the ADNI cohort including 3,123 SNPs and 2 types of biomarkers, VBM and FreeSurfer. Empirical results demonstrated advantages of our proposed model over the counterparts. Moreover, available literature was identified for our top selected SNPs, which demonstrated the rationality of our prediction results. An executable program is available online at https://github.com/littleq1991/sparse_lowRank_regression.

Learning Multi-Modal Biomarker Representations via Globally Aligned Longitudinal Enrichments

“…First, since AD is a progressive neurodegenerative disorder, multiple records can be obtained to monitor the disease's progression. Thus it is beneficial to uncover the temporal relation among these longitudinal biomarkers (Wang et al 2012c;Wang, Shen, and Huang 2016;Wang et al 2017;Brand et al 2018). Second, while heterogeneous biomarker measurements, such as voxel-based morphometry (VBM), FreeSurfer, and single-nucleotide polymorphism (SNP), are available for predicting disease status and/or cognitive performance, current longitudinal methods (Wang et al 2012a;Wang, Shen, and Huang 2016) often do not explore this multi-modal structure to boost prediction capabilities.…”

“…To overcome the first limitation of longitudinal data, several proposed longitudinal prediction models (Wang et al 2012c;Wang, Shen, and Huang 2016;Wang et al 2017;Lu et al 2018;Brand et al 2018) uncover the temporal structure of brain phenotypes. However, these models treat the temporal biomarkers as a tensor, which inevitably increases the complexity of the prediction problem.…”

Quantitative trait loci identification for brain endophenotypes via new additive model with random networks

“…Several machine learning models were established to depict the relations between SNPs and brain endophenotypes (Huo et al, 2018;Wang et al, 2012a;Wang et al, 2017;Yang et al, 2015;Zhu et al, 2016). In Wang et al (2012a), Zhu et al (2016), Wang et al (2017) and Huo et al (2018), the authors used the low-rank learning models or structured sparse learning models to select the imaging features that share common effects in the regression analysis. Yang et al (2015) applied the LASSO regression model to discover the significant SNPs that are associated with brain imaging features.…”