QFASA: A Comprehensive R Package for Diet Estimation via Fatty Acid Signature Analysis

Quantitative fatty acid signature analysis (QFASA) is a well-established diet estimation method that has been used extensively on a wide variety of marine mammal species. The method, along with its new refinements and extensions, requires the use of statistically intricate tools, many of which are computationally demanding. Recent developments in QFASA include a maximum likelihood framework for diet estimation, statistically valid inference procedures such as confidence intervals for the diet and hypothesis tests for comparing fatty acid signatures and/or diets, a measure of repeatability in the diet estimates, a prey species selection algorithm, as well as novel ways to estimate calibration coefficients, which are used to improve accuracy in the estimates. The QFASA R package was developed to facilitate access to the latest statistical QFASA tools and provide a means of efficiently disseminating new QFASA-related research, often developed by statisticians in collaboration with biologists. Further, using up-to-date functions ensures that QFASA methods are being applied in a legitimate and consistent manner. In this work, we present the QFASA R package, highlighting key functions for diet estimation and demonstrating their use with sample data available in the package. The QFASA R package is user-friendly, offers a broad range of functionality, and the vast majority of the functions are unique to this package.