Background

Biology has become a data-rich science. There are rapid developments of new technologies, and the data we can now gather about biological systems are large, heterogeneous and complex. We need to be able to  efficiently explore the data and draw sound conclusions in a transparent and reproducible manner. The field of statistics  increasingly makes use of, and is driven by, computational methods,  including resampling, simulations. As a result, exploratory data analysis and visualization, which are ever important, are becoming increasingly accessible.

Project

Modern Statistics for Modern Biology

The Holmes and Huber labs collaborate on developing statistical tools for large multi-layer data analyses,  for integrating large, heterogeneous biological data, and for finding applications in molecular medicine. We aim to deliver tools that are easy to use by domain-scientists to analyze their own data – for instance by providing the tools in the form of R / Bioconductor packages.

Together we want to help the next generation of biologists understand the “black box” of statistics by training them in quantitative statistical methods. We have written a textbook (Modern Statistics for Modern Biology) and together, we teach a summer course (Stats 366 – Bios 221) at Stanford. We keep further developing these materials, to take up new scientific developments (e.g.  new data types), new methods, or new statistical or computational ideas. 

Find out more:

Interested in developing computational tools for statistical analysis? Do you want to learn more about this great resource for biologists? Get in touch, we would love to hear from you!