Embedded Files
Date: Nov 9, 2018
Speaker: Guenter Tusch
Abstract
Genomic data become more frequently part of clinical practice. Novel tools and methods are required to transform information from increasingly voluminous genomic databases into actionable data for health care. In the era of precision medicine, the development of high-throughput technologies and electronic health records resulted in a paradigm shift in healthcare. However, the treatment of temporal data still remains a challenge. Recent efforts propose temporal models for the electronic health record, but not for genomic data.
One frequently employed model for temporal data in healthcare is temporal abstraction, a model based on conversion of expression values into an interval-based qualitative representation expressing the amount of change over time. The challenge is to find a domain specific mapping to create those representations. This study explores the feasibility of a hybrid AI-statistical model where the amount of change is determined by statistical significance the most reliable measure to determine biological significance. We propose to use empirical Bayes methods to determine differences in consecutive time points. Comparisons across platforms are done by comparing p-values. For DGE count data we use the voom transformation allowing RNA-seq data to be analyzed in a similar way. We demonstrate this approach in the framework of our SPOT software.
References
Google Sites
Report abuse