数学科学研究所
Insitute of Mathematical Science

Applied Mathematics Seminar 15: A Partially Functional Linear Modeling Framework for Integrating Genetic, Imaging, and Clinical Data

Seminar| Institute of Mathematical Sciences
Time:Friday, December 9th, 2022, 16:15-17:15
Location:RS408; Online, Tencent Meeting
Speaker: Ting Li, Shanghai University of Finance and Economics

AbstractThis paper is motivated by the joint analysis of  genetic, imaging, and clinical (GIC) data collected in many  large-scale   biomedical studies, such as the UK Biobank study and the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. We propose  a regression framework based on  partially functional linear regression models to  map high-dimensional GIC-related pathways for phenotypes of interest. We develop a joint model selection and estimation procedure by embedding imaging data in the reproducing kernel Hilbert space and imposing the $\ell_0$ penalty for the coefficients of scalar variables. We systematically investigate the theoretical properties of scalar and functional efficient estimators, including  non-asymptotic error bound,  minimax error bound, and asymptotic normality. We apply the proposed method to the ADNI dataset to identify important features from several millions of genetic polymorphisms and study the effects of a certain set of informative genetic variants and the hippocampus surface  on thirteen cognitive variables.

Tencent Meeting Number : 702-6472-0755

地址:上海市浦东新区华夏中路393号
邮编:201210
上海市徐汇区岳阳路319号8号楼
200031(岳阳路校区)