Graduate Student Seminar

Speaker 1: Gregory Haber

Title

An Introduction to Estimation of Binomial Parameters Using Group Testing

Abstract

Since group testing was introduced in 1943 it has found a wide range of application in medicine, biology, engineering, manufacturing, and beyond.  While orginally designed to make the detection of non-conforming members of a given population easier and cheaper, the methodology has also found an important application in the estimation of binomial prevalence. In this talk, we give an introduction to the field, including its roots in Dorfman's seminal work and the more modern estimation methods. We conclude with a brief description of potential research topics related to the estimation problem.

Speaker 2: Iris Gauran

Title

Empirical Null using Discrete Mixture Distributions and its Application to Protein Domain Data

Abstract

In recent mutation studies, analyses based on protein domain positions are gaining popularity than gene-centric approaches since the latter have limitations in considering the functional context that the position of the mutation provides.  This presents a large-scale simultaneous inference problem, with hundreds of hypothesis tests to consider at the same time.  This paper aims to select significant mutation counts while controlling a given level of Type I error via False Discovery Rate (FDR) procedures.  Simulated and protein domain data sets are used to illustrate a novel procedure in estimation of the empirical null using a mixture of discrete distributions and its use in local false discovery rate.