High-throughput sequencing view on the magnitude of global fungal diversity

Abstract

High-throughput DNA sequencing has dramatically transformed several areas of biodiversity research including mycology. Despite limitations, high-throughput sequencing is nowadays a predominant method to characterize the alpha and beta diversity of fungal communities. Across the papers utilizing high-throughput sequencing approaches to study natural habitats in terrestrial ecosystems worldwide, > 200 studies published until 2019 have generated over 250 million sequences of the primary mycological metabarcoding marker, the nuclear ribosomal internal transcribed spacer 2 (ITS2). Here we show that at a 97% sequence similarity threshold, the total richness of non-singleton fungal taxa across the studies published so far is 1.08 million, mostly Ascomycota (56.8% of the taxa) and Basidiomycota (36.7% of the taxa). The Chao-1 estimate of the total extant fungal diversity based on this dataset is 6.28 million taxa, representing a conservative estimate of global fungal species richness. Soil and litter represent the habitats with the highest alpha diversity of fungi followed by air, plant shoots, plant roots and deadwood with Chao-1 predictions, for samples containing 5000 sequences, of 1219, 569, 392, 228, 215 and 140 molecular species, respectively. Based on the high-throughput sequencing data, the highest proportion of unknown fungal species is associated with samples of lichen and plant tissues. When considering the use of high-throughput sequencing for the estimation of global fungal diversity, the limitations of the method have to be taken into account, some of which are sequencing platform-specific while others are inherent to the metabarcoding approaches of species representation. In this respect, high-throughput sequencing data can complement fungal diversity predictions based on methods of traditional mycology and increase our understanding of fungal biodiversity.