Online Program Home
My Program

Abstract Details

Activity Number: 410 - High-Dimensional Regression
Type: Contributed
Date/Time: Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #329419 Presentation
Title: Finding Needles in a Hay Stack - an Approach for a Small-Number-Factor High-Dimensional Data
Author(s): Chi-Hse Teng*
Keywords: high-dimensional data; small sample ; small-number-factor

We present a data analysis approach for a project that engineered mRNA constructs to max expression and half-life of a secreted protein. We selected 22 3'UTRs, 105 5'UTRs and 93 Signal peptides of highly expressed, secreted skeletal muscle proteins to test their relative influences on expression of the protein of interest. Given 22 3'UTRs, 105 5'UTRs and 93 Signal peptides, an all-encompassing experiment would have required 214,830 constructs to include all combinations. This would have been an overwhelming effort. The screen was limited to an examination of 218 constructs instead. The data analysis effort is to identify 3'UTRs, 5'UTRs and Signal peptides that might increase the expression of protein. We developed an approach to rank the sequences of each region. The approach delivered sensible results matching the experimental data of some known sequence's ranking. It also performed well in the simulation cases.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program