Skip Navigation
  • UC Riverside
  • University Advancement
  • Strategic Communications

Testing Reproducibility of Results Across 'omics Studies

Loki Natarajan, Ph.D., Adjunct Professor, University of California, San Diego

Tuesday, March 14, 2017
  3:45–4:45 p.m.

Location: Olmsted Hall 420
  Parking Information

Category: Lecture


Loki Natarajan, Ph.D.
Adjunct Professor
Department Family Medicine and Public Health
University of California, San Diego

Testing Reproducibility of Results Across 'omics Studies

Curated public repositories of 'omics data are becoming increasingly available. Such databases enable researchers to compare results across multiple 'omics studies in order to replicate findings.  A common approach is to first rank "genes" (or features) according to a hypothesis of interest within each study. Then, lists of the top-ranked genes within a study are compared across studies.  Genes recaptured as highly ranked (usually above some threshold) in multiple studies are considered to be significant.  In this talk, we develop a formal inferential strategy for this kind of list-intersection discovery test.  We will show how to compute a p-value associated with a `recaptured' set of genes, using a closed-form Poisson approximation to the distribution of the size of the recaptured set.   We will investigate operating characteristics of the test as a function of the total number of studies considered, the rank threshold within each study, and the number of studies within which a gene must be recaptured to be declared significant. We give practical guidance on how to design a bioinformatic list-intersection study with adequate control of Type I error and false discovery rate, while maximizing expected sensitivity to capture true positive genes. We will illustrate our methods using prostate cancer gene-expression datasets from the curated Oncomine database. (This is joint work with Karen Messer, Minya Pu)

Dr. Natarajan received her PhD in Mathematics in 1991 from UC Berkeley. Spurred by exciting developments in computational biology, she retooled in biostatistics at UCSD. She was appointed Assistant Professor of Biostatistics at the UCSD School of Medicine in 2002, and promoted to Professor in 2011. Dr. Natarajan is PI (co-PI Karen Messer) of a NIH-funded study examining genomic, clinical and behavioral factors prognostic for breast cancer survival.  She is an active member of the Moores UCSD Cancer Center and the UCSD Institute for Metabolic Medicine, and is co-investigator on multiple studies on health behaviors, obesity and chronic disease, cancer prevention, and diabetic kidney disease. Dr. Natarajan is also Program Director for the UCSD PhD program in Biostatistics, which was launched in Fall 2016. Dr. Natarajan's primary methodological research interests are in the area of exposure measurement error, with particular emphasis on dietary self-report. She is also working on efficient designs for Phase I trials, developing models for survival data when covariates violate the proportional hazards assumption of the usual Cox model, mutation rate estimation in cancer cell lines, and analysis of high-dimensional genomic data.
Dr. Natarajan is involved as lead biostatistician in many collaborative projects. A sampling of these include the Women's Health Eating, Living (WHEL) clinical trial and survivorship study (PI John Pierce), a Transdisciplinary Research in Energetics and Cancer (TREC) Program Project (PI Ruth Patterson), and several clinical and observational studies of sleep and inflammation (PIs Sonia Ancoli-Israel, Joel Dimsdale, Paul Mills). She is a member of the Biostatistics and Bioinformatics Shared Resource of the Moores UCSD Cancer Center (Dr. Karen Messer, Director), a statistical consulting service that provides statistical support to Cancer Center members.

Open to: General Public
Admission: Free
Sponsor: Statistics Department

Contact Information:
Paula Lemire
(951) 827-6009