- This event has passed.
Learn About R Notebooks, aheatmap Package and Interactive Graphics
February 20, 2017 @ 5:30 pm UTC+0
R-Ladies Melbourne is pleased to announce the first event of the year. We will be hosting two amazing women! Let’s learn more about life-saving features of R together while networking and helping each other. Snacks and beverages will be provided.
Here are the details of the event.
Date & Time: Monday, February 20, at 5:30pm
Location: Walter and Eliza Hall Institute, Level 7, Davis Auditorium
Speakers: Jovana Maksimovic and Natalia Da Silva
Jovana Maksimovic After completing a Bachelor of Science (Honours) / Bachelor of Bioinformatics at La Trobe University, majoring in biochemistry, genetics and computer science, Dr Maksimovic worked at the Department of Primary Industries for two years as part of their graduate program. During this time, she worked on many diverse research projects and developed an interest in the biology of lactation. She began her PhD at Monash in 2007 on a Department of Primary Industries project investigating the expression and regulation of a gene family involved in the production of a subset of milk oligosaccharides that are of interest in infant nutrition. She currently works as a Postdoctoral Scientist in the bioinformatics group at Murdoch Childrens Research Institute. Dr Maksimovic currently works on the analysis of epigenetic data, with a focus on methylation, particularly in immune cells.
Title: Some of my new favourite things for working with R
Abstract: As a bioinformatician, I analyse different types of high-dimensional data to answer biological questions. My work requires a lot of data exploration and is often highly iterative. R markdown has become indispensable for keeping my projects organised, reproducible and maintaining version control. However, when working with large datasets involving lots of plots, it was tedious to have to render an entire document to see the results of changes to a single plot. Alternatively, each plot had to be refined on its own, which was acceptable but could become error prone when preceding code was changed. The recent introduction of R notebooks in RStudio 1.0.0 has revolutionised how I work with R markdown. The ability to view plots/results in the source has made editing easy and robust, whilst retaining all the reproducibility and version control perks of traditional R markdown. My recent discovery of the aheatmap function in the NMF CRAN package has also made my life a lot easier by drastically simplifying the code require to create an informative heatmap.
Natalia Da Silva is a Ph.D. candidate in Statistics at Iowa State University working with Dianne Cook and Heike Hofmann. She completed her M.S. in Statistics at Iowa State University working in network meta-analysis with Bayesian inference in Spring 2014 and received a double B.S. in Economics (Spring 2009) and in Statistics (Spring 2008) at University of the Republic (Montevideo, Uruguay). Her Ph.D. thesis topic is bagged projection methods for supervised classification in big data. She is interested in data mining, prediction, exploratory data analysis, statistical graphics, reproducible research and meta-analysis.
She enjoys being outdoors and her favorite season is summer, therefore she is happy to be in Australia then.
Title: Interactive graphics for visually diagnosing forest classifiers in R.
Abstract: This talk describes structuring data and constructing plots to explore forest classification models interactively. A forest classifier is an example of an ensemble, produced by bagging multiple trees. The process of bagging and combining results from multiple trees, produces numerous diagnostics which with interactive graphics can provide a lot of insight into class structure in high dimensions. Various aspects are explored in this talk, to assess model complexity, individual model contributions, variable importance and dimension reduction, and uncertainty in prediction associated with individual observations. The ideas are applied to the random forest algorithm, and to the projection pursuit forest, but could be more broadly applied to other bagged ensembles. Interactive graphics are built in R, using ggplot2, plotly, and shiny packages.
Sign up now!!
We are looking forward to seeing you,