Original ArticlesFrom phylogenetic to functional originality: Guide through indices and new developments

- A new index of species' functional originality is presented and evaluated.
- The properties of this index and a range of published indices are analyzed.
- Phylogenetic originality can rarely serve as a proxy for functional originality.
- Transforming functional data into functional trees modifies species' originalities.
- Indices are extended to originality at all scales from populations to regions.

In biodiversity studies a species is often classified as original when it has few closely related species, a definition that reflects its phylogenetic originality. More recently, studies have focussed on biological or functional traits that reflect the role(s) that species play within communities and ecosystems. This has led many studies to an alternative evaluation of species' originality: its functional originality. Most indices of species' originality were developed to treat the hierarchical structure of a (phylogenetic) tree. The change in perspective from measures of phylogenetic originality to measures of functional originality thus raises methodological issues particularly around the need to develop indices explicitly appropriate for evaluating functional trait-based originality. We compare indices of species' originality including a new index which we develop to evaluate (1) whether phylogenetic originality could serve as a proxy for functional originality in conservation and ecological studies; (2) whether the transformation of functional data into functional trees modifies the way species are ranked according to their originality measures compared to approaches that directly rely on pairwise functional dissimilarities among species; and more generally, (3) whether different indices provide different views on how original species are from each other, hence reflecting different ecological and evolutionary processes that generated patterns of originality. Using simulations and a real case study, we show that: (1) the strong effects of the choice of a clustering approach can affect reported levels of dissimilarities among species; (2) the tree-based approaches could better reflect the trait-generating processes under constant (Brownian) rates of evolution; and (3) phylogenetic originality measures can depart from functional originality measures when species have large amount of independent evolution. Overall, phylogenies may be used at large scales but cannot replace functional approaches designed for depicting community assembly. Indeed, traits involved in ecological processes may have various histories and thus moderate phylogenetic signals. Our comparative study provides approaches and perspectives on the analysis of originality across biological scales of organization from individuals, through populations, up to the originalities of communities and regions.