Graphical Models for Differential Gene Expression

Lunch will begin at 11:45am

ABSTRACT:
 

Anomalies in gene expression can serve as important diagnostic signals in cases of undiagnosed genetic disease.  In the case of allelic imbalance in expression, existing methods either fail to make use of all available information, or utilize heuristics that result in unwanted biases or lack of statistical power.  In this work we develop novel approaches to integrating short-read RNA sequencing data across multiple heterozygous sites while accounting for uncertainty in phasing, in a principled, probabilistic manner, and show that this results in higher quantification accuracy for allele-specific expression in endogenous genes.  We also consider the case of mutations tested out-of-genome in massively parallel reporter assays, and show that a similar modeling approach allows detection of regulatory effects at lower allele frequencies than previously possible.  Finally, we show that data from these assays can be used to build predictive models to impute regulatory effects of untested variants, based on local sequence alone.

BIO:

Dr. Majoros is an Assistant Professor in the Department of Biostatistics and Bioinformatics at Duke.  He graduated magna cum laude from Penn State in 1994, earning a bachelor's degree in Computer Science, and in 2017 completed a PhD in Computational Biology and Bioinformatics in the CBB program at Duke.  

He has been working as an applied researcher for over 25 years, in a range of areas including defense electronics, natural language processing, and genomics.  In 2000 he contributed to the initial sequencing and annotation of the human genome at Celera Genomics.  That work culminated in his seminal book on algorithms for gene structure prediction, which was published by Cambridge University Press in 2007.

He has been at Duke continuously for over 20 years, initially as staff, then as PhD student, then postdoc, and now as faculty. During that time, he has focused on developing novel computational methods for use in understanding genetic mechanisms in gene regulation.