Statistical Inference in mRNA-Seq
Mathematical Genomics
April 13, 2009 07:00 PM to 07:45 PM
Speakers:
Dudoit, Sandrine
|
 |
Abstract: |
For the past decade, microarrays have been the assays of choice for
high-throughput studies of gene expression. Recent improvements in
the efficiency, quality, and cost of genome-wide sequencing are
prompting biologists to rapidly abandon microarrays in favor of ultra
high-throughput sequencing, a.k.a., second-generation or
next-generation sequencing: e.g., Applied Biosystems' SOLiD, Helicos
BioSciences' HeliScope, Illumina's Genome Analyzer, and Roche's 454
Life Sciences sequencing systems. These high-throughput sequencing
technologies have already been applied to monitor genome-wide
transcription levels (mRNA-Seq), DNA-protein interactions (ChIP-Seq),
chromatin structure, and DNA methylation. While
sequencing-based gene expression studies have been touted as
overcoming longstanding limitations of microarray-based studies, these
new biotechnologies raise similar as well as novel statistical and
computational challenges, in areas such as image analysis,
base-calling, read-mapping, and (differential) expression inference.
This talk will report on our investigation of two mRNA-Seq datasets
obtained using Illumina's Genome Analyzer platform to measure
transcript levels in reference samples from the MicroArray Quality
Control (MAQC) Project. We focus on the analysis of mapped read counts
and the following three main issues: (1) exploratory data analysis
(EDA); (2) assessment of biological effects of interest (e.g.,
expression levels in Brain vs. UHR RNA) and nuisance experimental
effects (e.g., library preparation, flow-cell, and lane effects); (3)
identification of differentially expressed genes.
This is joint work with James H. Bullard, Steffen Durinck, Kasper
D. Hansen, and Elizabeth A. Purdom. |
|
|
