|
Activity Number:
|
81
- Contributed Poster Presentations: Section on Statistics in Epidemiology
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Section on Statistics in Genomics and Genetics
|
|
Abstract #313121
|
|
|
Title:
|
Gene Network Construction Using Ordering Information
|
|
Author(s):
|
Meichen Dong* and Fei Zou
|
|
Companies:
|
University of North Carolina at Chapel Hill and University of North Carolina at Chapel Hill
|
|
Keywords:
|
gene regulatory network;
fused-lasso;
scRNA-seq;
spatial transcriptomics
|
|
Abstract:
|
Integrating biological knowledge with gene regulatory networks and pathways can shed light on therapeutic targets of complex human diseases. Cutting edge single-cell sequencing data enables researchers to study networks and pathways for complex tissues and under different conditions. Current methods treat single cells equally distant from each other under a condition without exploiting the inter-cell relationships. Here, we propose a framework to construct gene networks based on fused-lasso regression where single cells are first ordered after dimension reduction, and then the ordering information is used to induce constraints on the covariance structures among the single cells to construct more robust and efficient networks. The proposed method is first illustrated on synthetically simulated data and then applied to a medulloblastoma scRNA-seq dataset and a spatial transcriptomic dataset.
|
Authors who are presenting talks have a * after their name.
