→ Slides
→ Video (
YouTube,
Vimeo)
Abstract:High throughput sequencing (HTS) has become a preferred choice for the measurement of genome-wide biological phenomena at the molecular level, from genetics to gene expression regulation. Despite its widespread use, challenges remain in HTS data analysis. One often-overlooked aspect is normalization. Despite the fact that a variety of factors or “confounders” can contribute unwanted variation to the data, commonly used normalization methods often only correct for sequencing depth. The study of gene expression and gene expression regulation is particularly problematic when it is influenced simultaneously by a variety of biological factors in addition to the one of interest. Using examples from our lab and others, we show that confounders can dominate the signal of interest. If the effect of confounders is not properly accounted for the power, reproducibility and biological insight of the results is compromised. I will focus on two methods aimed at correcting for confounding factors: RUVSeq (for RNA-seq) and DEScan (for epigenomic-seq). Our results show that removing confounding factors can make a dramatic difference on the biological conclusions of genome-wide studies of gene expression and epigenomic regulation.
Dr. Peixoto is an Assistant Professor at the
Elson S. Floyd College of Medicine at
Washington State University. Her research focuses on using genomic and computational biology approaches to study brain function. She also advises the
WSU Spokane Genomics Core, and has published work on how analysis of complex datasets affects power and reproducibly in RNA-seq and Epigenomic-seq.
Lucia's keynote is sponsored by GCC.