In statdivlab/CatchMore: Species Richness estimation with CatchAll

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "README-"
)

CatchMore

CatchMore is an R package to estimate species richness with mixed-Poisson and non-mixed-Poisson models. It extends the historically popular methodology CatchAll to include a more diverse catalogue of models, and provides an easy-to-use R implementation that runs on Unix machines (unlike CatchAll).

CatchMore is written by Kendrick Qijin Li and Amy Willis, and is maintained by the Statistical Diversity Lab at the University of Washington.

Installation

devtools::install_github("statdivlab/CatchMore")
library(CatchMore)

Basic Usage

Let's illustrate CatchMore using the apples dataset from the breakaway package

library(breakaway)
data(apples)

CatchMore can run a number of different exponential-mixed Poisson models, including the Poisson model (a zero-exponential mixture), the geometric model (a single exponential-mixed Poisson model), and two- or three-component exponential mixture:

Poisson_model(apples, cutoff = 20)
geometric_model(apples, cutoff = 20)
two_geometric_model(apples, cutoff = 50)
three_geometric_model(apples, cutoff = 70)

To pick between them, we implemented catch_all, which produces the same results as legacy CatchAll, and our new model selection procedure catch_more:

catch_all(apples)
catch_more(apples)

Citing CatchMore

Writing software is time-consuming and often thankless. If you use this software, please cite our work in the following way:

Li, Q.K. and Willis, A.D. (2019). CatchMore: an R package for richness estimation. In Preparation.

CatchMore implements a number of different richness estimates. Please also cite the following if you use them:

• breakaway() and kemp(): Willis, A. & Bunge, J. (2015). Estimating diversity via frequency ratios. Biometrics.
• catch_all(): Bunge, J. (2011). Estimating the number of species with CatchAll. Proceedings of the Pacific Symposium on Biocomputing.

statdivlab/CatchMore documentation built on May 8, 2019, 8:12 a.m.