I am passionate about the application of statistical methods in fields of research that improve human health. I have had extensive experience in the application and development of statistical methods for the analysis of medical and scientific data as a Senior Statistical Analyst in Northwestern University’s Department of Preventive Medicine, where I have worked since completing my M.S. in epidemiology and biostatistics in 2013. This experience, along with my quantitative coursework, has motivated me to pursue a doctoral degree in biostatistics. In pursuing a doctoral degree, I wish to deepen my existing knowledge of statistical theory and methodology in order to take on positions of leadership in the design, conduct, and analysis of collaborative …show more content…
My coursework and research experiences in the biological sciences have aided me in my current position, and will aid me in a career as a collaborative biostatistician, by allowing me to understand the unique features of various types of medical and scientific data, and to put those data in context. In addition to scientific coursework, I have worked to complete quantitative and computational coursework that will prepare me for the demands of a doctoral program. During the course of my Master’s program, I excelled in courses on biostatistics, calculus-based probability, statistical inference, data mining in R, and data analysis and management in SAS. I have also completed coursework in single and multivariable calculus as part of a post-baccalaureate certificate in mathematics from Northwestern University, in which I have a 4.0 GPA, and I will complete a course in linear algebra by spring 2016. In addition to my participation as a student, I have also had the opportunity to teach biostatistics to health professionals, research scientists, and graduate students as the Teaching Assistant for Introduction to Biostatistics for three …show more content…
One of my recent responsibilities has been the analysis of gas chromatography/mass-spectrometry (GC/MS) metabolomics data from the Hyperglycemia and Adverse Pregnancy Outcomes (HAPO) study. These data feature a high proportion of missingness due to true metabolite absence, as well as metabolite presence below levels of technical detectability. In addition, these data are subject to substantial technical variability, including metabolite-specific batch and run-order effects. This presents a substantial challenge for large-scale studies like HAPO, where samples must be run in different batches over several weeks. In order to effectively analyze these data, I collaborated on the development of metabomxtr, an R package which facilitates mixture-model analysis of GC/MS metabolomics data and explicitly models metabolite presence/absence. I then extended the use of the mixture-model to a normalization context. I am the first author of a submitted manuscript detailing the use of metabomxtr for normalization purposes, and a primary contributing author of the method’s initial publication for