Title: "Statistical calibration of high-throughput gene expression data using measurement error models"
Speaker: Zhaonan Sun, Department of Statistics, Purdue University

Place: SMITH (SMTH) Hall 108
Date: March 4, 2014; Tuesday
Time: 4:30pm


Accurate quantification of gene expression is crucial for any transcriptome study. In the last decade microarray technology has been used as a main platform for simultaneous interrogation of thousands of genes. Recently, next-generation sequencing technologies have emerged as promising alternatives to microarrays; when applied to transcriptomes it is known as RNA-seq. Unfortunately, like microarray expression data, RNA-seq data are subject to various measurement errors. Using a third platform, called real-time reverse-transcription polymerase chain reaction (qRT-PCR), which has limited throughput capacity, but provides more accurate quantification of gene expression than either microarray or RNA-seq technologies, we propose a system of functional measurement error models that both models gene expression measurements and calibrates the microarray and RNA-seq platforms to qRT-PCR. Both theoretical and simulation studies were conducted to establish the properties of the calibrated expression measurements. The model and approach were applied to data from the Microarray Quality Control (MAQC) and Sequencing Quality Control (SEQC) projects; these results will be discussed.

Associated reading:
Zhaonan Sun, Thomas Kuczek, and Yu Zhu (2014): Statistical calibration for qRT-PCR, microarray and RNA-Seq expression data with measurement error models. To appear in Annals of Applied Statistics. arXiv link: http://arxiv.org/abs/1212.6690

Click here www.stat.purdue.edu/~doerge/BIOINFORM.D/SPRING14/sem.html for a full schedule of BIOINFORMATICS SEMINARS, past and present.