Paper introducing bayesTPC is out in Methods in Ecology and Evolution

The QED Lab at Virginia Tech is very pleased to announce that the paper introducing bayesTPC has appeared in Methods in Ecology and Evolution. The bayesTPC R package is for Bayesian fitting of Thermal Performance Curves (TPCs) to trait responses. The package is built using the nimble language and machinery as the underlying engine for Markov Chain Monte Carlo (MCMC). Our main aims in releasing bayesTPC were to support the adoption of Bayesian approaches in thermal physiology, and to promote TPC fitting that adequately quantifies uncertainty.

The paper, led by former QED undergraduate researcher Sean Sorek, introduces the basic functionality of the package. It takes users through installation and a few basic analyses, including methods for model checking and comparison. One of the exciting features of the package is that it has built-in functions that allow a user to pull data in the VecTraits database directly into R. In the paper we use this feature explicitly, so that readers can see how data from VecTraits may be analyzed with the bayesTPC package.

We've been working on the package for a few years, testing the functionality and usability through a VectorByte training workshop and an EEID training workshop. Thus, in addition to the guidance in the paper itself, we have two sets of training materials (using some older versions of the package), and an additional set of tutorials have been developed by Sean. We plan to create an updated version of the tutorials over the summer and link them to the rest of the VectorByte resources - so stay tuned!