PT18 – Complementary Feature Selection from Alternative Splicing Events and Gene Expression for Phenotype Prediction

Proceeding talk – Theme: Data.

Abstract

A central task of bioinformatics is to develop sensitive and specific means of providing medical prognoses from biomarker patterns. Common methods to predict phenotypes in RNA-Seq datasets utilize machine learning algorithms trained via gene expression. Isoforms, however, generated from alternative splicing, may provide a novel and complementary set of transcripts for phenotype prediction. In contrast to gene expression, the number of isoforms increases significantly due to numerous alternative splicing patterns, resulting in a prioritization problem for many machine learning algorithms. This study identifies the empirically optimal methods of transcript quantification, feature engineering, and filtering steps using phenotype prediction accuracy as a metric. At the same time, the complementary nature of gene and isoform data is analyzed and the feasibility of identifying isoforms as biomarker candidates is examined.

Link to PDF file

Authors

Charles Labuzzetta, Iowa State University, United States
Margaret Antonio, Boston College, United States
Patricia Watson, Medical University of South Carolina, United States
Robert Wilson, Medical University of South Carolina, United States
Lauren Laboissonniere, Iowa State University, United States
Jeffrey Trimarchi, Iowa State University, United States
Baris Genc, Northwestern University Feinberg School of Medicine, United States
P. Hande Ozdinler, Northwestern University Feinberg School of Medicine, United States
Dennis Watson, Medical University of South Carolina, United States
Paul Anderson, College of Charleston, United States