A metagenomic alpha-di v ersity index for microbial functional biodi v ersity

Abstract

Alpha-di v ersity indices are an essential tool for describing and comparing biodiversity. Microbial ecologists apply indices originally intended for, or adopted by, macroecology to address questions relating to taxonomy (conserved marker) and function (metagenome- based data). In this Perspecti v e piece , I be gin by discussing the natur e and mathematical quirks important for interpr eting r outinel y employed alpha-di v ersity indices. Secondl y, I pr opose a meta genomic alpha-di v ersity index ( M D ) that measures the (dis)similarity of protein-encoding genes within a community. M D has defined limits, whereby a community comprised mostly of similar, poorly di v erse pr otein-encoding genes pulls the index to the lower limit, while a community rich in di v erg ent homologs and unique g enes dri v es it tow ard the upper limit. With data acquir ed fr om an in silico and three in situ meta genome studies, I deri v e M D and typical alpha-di v ersity indices applied to taxonomic (ribosomal rRNA) and functional (all protein-encoding) genes, and discuss their relation- ships with each other. Not all alpha-di v ersity indices detect biological trends, and taxonomic does not necessarily follow functional biodi v ersity. Thr oughout, I explain that protein Richness and M D provide complementary and easily interpreted information, while pr oba bility-based indices do not. F inally, consider ations re garding the unique nature of microbial metagenomic data and its relevance for describing functional biodi v ersity ar e discussed.