Project Description

Dr Nicola Armstrong

Murdoch University

Classification and prediction ‘omics style

Data science and machine learning for bioinformatics

Tuesday 3 July 2018

Dr Armstrong is a statistical bioinformatician who completed her doctoral studies in Statistics at the University of California, Berkeley. After graduating with her PhD, she spent several years in the Netherlands as a postdoc at Eurandom and the Vrije Universiteit before moving to the Netherlands Cancer Institute in Amsterdam as a senior statistician. On returning to Australia, she worked at the Garvan Institute and the University of Sydney before moving to Murdoch University where she is currently a Senior Lecturer in mathematics and statistics. Her research work has centered on the development of statistical methodology and the application of statistics to problems in genetics, genomics and biomedical research.

In this talk, we will explore different classification and prediction methods that are commonly used on ‘omics data, and discuss the advantages and disadvantages of the various approaches. Quantifying the performance of a classifier and other important concepts that should be considered when approaching a classification problem will also be explored.