CGSA Workshops have begun!

The new academic year is off to a great start and we’ve now had two CGSA Chats or Workshops.

Photo Credit: Catherine Herzog

On September 7^th, Dr. Maciej Boni met with the graduate students for a whiteboard chat about phylogenetics and the questions ‘What is a phylogenetic tree?’, ‘What are the types of phylogenetic trees?’, and ‘How are phylogenetic trees built?’

Maciej reviewed clustering, k-means, unweighted pair group mean algorithm (UPGMA), neighbor joining approach, parsimony trees, max-likelihood trees, and multiple models of evolution named after scientists that have accumulated over the years.  He continued to explain different approaches to tree building and highlighted that some approaches have flaws when considering evolution.  One example was UPGMA, which does not handle long branches well.  Additional topics discussed included how to root trees, the concept of a molecular clock, and which different software tools were used in practice when building these trees.  There will be a follow-up CGSA Chat with Maciej on Thursday September 28 at 3:45pm in W201 MSC which will focus more on a specific software and doing practicals in tree building.

The second workshop, led by post-doc Dr. Amalie McKee and held on September 15^th, was a hands-on session titled ‘Figure Troubleshooting’.  Graduate students brought figures they were working on or questions about the best ways to visualize certain types of data.  Amalie provided excellent advice regarding figures and data visualizations, which are summarized here:

• If you can’t draw a figure by hand, then it is too complicated. The best figures are ones you are able to draw.
• A model is a storytelling mechanism. Before you write one down, you should know the story you want to tell.  Same with figures.
• It may be a good idea to avoid 3D plots, as they often have more than three dimensions! Make sure to carefully consider the number of dimensions you are working with or trying to show.
• Remember that a contour plot is a 3D plot. You can iterate a contour plot over time to make it 4D.
• The best tool for making videos, such as stringing a series of figures produced in R together, is ImageMagik. This program works best if the figures (frames) are already generated as plots in R with consecutive numbering so it can string them together into a video.

Additionally, Amalie brought several books by Edward Tufte for the attendees to review.  These included: The Visual Display of Quantitative Information (1983), Envisioning Information (1990), Visual Explanations (1997), and Beautiful Evidence (2006).

 Information on upcoming workshops will be shared in the CIDD Weekly Digest emails.