canaper: Categorical analysis of neo- and paleo-endemism in R

Abstract

Biodiversity has typically been quantified using species richness, but this ignores evolutionary history. Due to the increasing availability of robust phylogenies, methods have been developed that incorporate phylogenetic relationships into quantification of biodiversity. CANAPE (categorical analysis of neo- and paleo-endemism) is one such method that can provide insight into the evolutionary processes generating biodiversity. The only currently available software implementing CANAPE is Biodiverse, which is written in Perl and can be used either through a graphical user interface (GUI) or user-developed scripts. However, many researchers, particularly in the fields of ecology and evolutionary biology, use the R programming language to conduct their analyses. Here, we present canaper, a new R package that provides functions to conduct CANAPE in R. canaper implements methods for efficient computation, including parallelization and encoding of community data as sparse matrices. The interface is designed for maximum simplicity and reproducibility; CANAPE can be conducted with two functions, and parallel computing can be enabled with one line of code. Our case study shows that canaper produces equivalent results to Biodiverse and can complete computations on moderately sized datasets quickly (< 10 min to reproduce a canonical study). canaper allows researchers to conduct all analyses from data import and cleaning through CANAPE within R, thereby averting the need to manually import and export data and analysis results between programs. We anticipate canaper will become a part of the toolkit for analyzing biodiversity in R.

Ancestral area analyses reveal Pleistocene‐influenced evolution in a clade of coastal plain endemic plants

“…The ASTRAL‐III phylogeny of the SMC and geographical data were imported into the ‘canaper’ package (Nitta, 2022), allowing us to compute raw biodiversity metrics using the ‘phyloregion’ package (Daru et al, 2020) and to run randomizations using the ‘picante’ package (Kembel et al, 2010), adding our spatial data into 20‐km grid cells in the process. Our grid grain size is based on the recommendations of Rahbek (2005), who suggested the use of a grain size comparable to the smallest range size among the species in the study area.…”

Correlated evolution in an ectomycorrhizal host–symbiont system

“…5) showed a trend that sites dominated by the older host species from subgenus Lophozonia (c. 72 Ma) contained fungal communities with longer phylogenetic branches than the younger hosts in the Fuscospora subgenus (c. 53 Ma). Whilst this result was marginally nonsignificant, the pattern was relatively striking, and the randomisation approach used is known to lack power when data come from less than five groups (Nitta et al, 2021). We also found a trend that Lophozonia-dominated sites had greater MPD compared with Fuscospora-dominated sties (Fig.…”

“…53 Ma). Whilst this result was marginally nonsignificant, the pattern was relatively striking, and the randomisation approach used is known to lack power when data come from less than five groups (Nitta et al ., 2021). We also found a trend that Lophozonia ‐dominated sites had greater MPD compared with Fuscospora ‐dominated sties (Fig.…”

“…A value > 1 indicates the community has an overrepresentation of long branches, and < 1 indicates greater prevalence of short branches than would be expected if the same number of taxa were randomly selected from the tree (Mishler et al ., 2014; Thornhill et al ., 2016). We calculated these statistics and performed null‐hypothesis randomisation tests to calculate z scores using the ‘cpr_rand_test’ function from the canaper R package (Nitta et al ., 2021) using the ‘independentswap’ null model with 100 000 iterations. Positive or negative z scores indicate that, for phylogenetic diversity, fungal ASVs associating with each host taxa are either more or less related than expected by chance, and for relative phylogenetic diversity, that the branch lengths of communities are either longer or shorter than expected by chance.…”

Global analysis of Poales diversification – parallel evolution in space and time into open and closed habitats

“…We used the R packages picante v.1.8.2 (Kembel et al ., 2010) and canaper v.0.0.2 (Nitta & Iwasaki, 2021) to calculate phylogenetic diversity (PD; Faith, 1992) and phylogenetic endemism (PE; Rosauer et al ., 2009), as well as to perform categorical analysis of neo- and palaeo-endemism (CANAPE; Mishler et al ., 2014) across botanical countries. To calculate the standardised effect size of PD and PE, we used two different sequential randomization algorithms – swap and curveball.…”