|
Activity Number:
|
436
- Network Inference for Omics and Imaging Data
|
|
Type:
|
Topic-Contributed
|
|
Date/Time:
|
Thursday, August 12, 2021 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #317650
|
|
|
Title:
|
Generalized Tensor Canonical Correlation Analysis for Network Inference Using Multi-Omics Data
|
|
Author(s):
|
Katerina Kechris* and Weixuan Liu and Farnoush Banaei-Kashani
|
|
Companies:
|
University of Colorado Anschutz Medical Campus and University of Colorado Anschutz Medical Campus and University of Colorado Denver
|
|
Keywords:
|
omics;
networks;
canonical correlation;
tensor
|
|
Abstract:
|
Complex diseases are often characterized by continuous phenotypic traits. However, many multi-omics analysis methods either do not incorporate traits, or only consider binary traits. We developed a generalized tensor canonical correlation analysis (GTCCA) approach for integrating multiple -omics data types along with a quantitative phenotype of interest to construct multi-omics networks specific to the phenotype. We used simulations to assess the ability of GTCCA to capture higher-order relationships among -omics datasets compared to non-tensor based approaches. As a case study, we applied this approach to a data set on chronic obstructive pulmonary disease to infer multi-omics networks associated with airflow obstruction. In addition to the flexibility of handling multiple -omics data sets, the GTCCA method can be easily generalized to incorporate multiple quantitative phenotypes simultaneously.
|
Authors who are presenting talks have a * after their name.
