Apple Bioinformatics Award Winners

Dr. Michael Newton

Left to Right: Dr. Michael Newton, Dr. Bret Larget, Dr. Sunduz Keles, Dr. Christina Kendziorski, Meng Chen (Grad Student), Mike Redmond (IT Services)

Institution

University of Wisconsin, Madison

Assignment

Professor, dual appointment with the departments of Statistics and of Biostatistics and Medical Informatics

Scientific Discipline

Statistical genomics

Challenge

To improve the discovery of biological signals by studying the statistics of genomic data.

Discoveries

First definitive support for bootstrapping in phylogenetics.

One of the first implementions of Markov chain Monte Carlo for Bayesian phylogenetic analysis.

The first mixture-model approach to measure differential gene expression from microarray data.

Research

“New technologies, gene expression arrays and chip technologies are creating enormous amounts of data for biologists. But the data includes all kinds of complicated noise. If you get enough noise, you can find patterns just by chance — like staring at the stars and seeing cups and saucers and faces. Precise statistics on the front end of data analysis can save researchers from chasing too many wild geese.”

“At the end of the day, an experiment generates a data file full of numbers. Were you somehow able to repeat the experiment, you’d almost certainly get different numbers. What are the reasons for that? What are the sources of variation? How can you accommodate different sources of variation in efforts to identify biological signals? We’re all about decomposing and analyzing variation.”

Resources

Newton’s team includes Drs. Douglas Bates, Sunduz Keles, Christina Kendziorski, and Bret Larget — experts in computational statistics, genomics, transcription regulation, microarray analysis, gene mapping and phylogenetics. Collaboration with scientists both in academia and industry also allow Newton’s team to develop innovative bioinformatics tools and statistical methods for analyzing emerging data.

Apple Workgroup Cluster for Bioinformatics

“Tools included with the Apple Workgroup Cluster are valuable for our for our research. We’ll also expand on the iNquiry tools for our pioneering research. Much of this work involves R, the widely used open-source statistical analysis system, and Bioconductor, the R-based bioinformatics collection of tools. Dr. Bates in our group is a leader in the core development of R.”

Why Apple?

“The Apple Workgroup Cluster offers the combination of speed, memory, storage and cost-effective expandability needed to handle the applications and algorithms that are essential to our research.”