What Is the Best Lens? Comparing the Resolution Power of Genome-Derived Markers and Standard Barcodes

Abstract

Fungal species delimitation was traditionally carried out with multicopy ribosomal RNA (rRNA) genes, principally for their ease of amplification. Since the efficacy of these markers has been questioned, single-copy protein-encoding genes have been proposed alone or in combination for Multi-Locus Sequence Typing (MLST). In this context, the role of the many sequences obtained with Next-Generation Sequencing (NGS) techniques, in both genomics and metagenomics, further pushes toward an analysis of the efficacy of NGS-derived markers and of the metrics to evaluate the marker efficacy in discriminating fungal species. This paper aims at proposing MeTRe (Mean Taxonomic Resolution), a novel index that could be used both for measuring marker efficacy and for assessing the actual resolution (i.e., the level of separation) between species obtained with different markers or their combinations. In this paper, we described and then employed this index to compare the efficacy of two rRNAs and four single-copy markers obtained from public databases as both an amplicon-based approach and genome-derived sequences. Two different groups of species were used, one with a pathogenic species of Candida that was characterized by relatively well-separated taxa, whereas the other, comprising some relevant species of the sensu stricto group of the genus Saccharomyces, included close species and interspecific hybrids. The results showed the ability of MeTRe to evaluate marker efficacy in general and genome-derived markers specifically.

When barcoding fails: Genome chimerization (admixing) and reticulation obscure phylogenetic and taxonomic relationships

“…The secondary barcodes used in mycology are conserved fragments of certain protein‐coding household genes usually present in single copies in haploid genomes (e.g., actin [ ACT1 ], β‐tubulin II [ TUB2 ], DNA‐directed RNA polymerase II largest [ RPB1 ] and second largest [ RPB2 ] subunits, translational elongation factor 1α [ TEF1 ], translation elongation factor 2 [ TEF2 / EF2 ] and calmodulin [ CAM ]) (Stielow et al, 2015). However, with the rapidly expanding amount of barcoding results and the increasing number of sequenced genomes, the accuracy and precision of taxon delimitation and taxonomic identification of strains by a handful of barcodes are increasingly being questioned (examples in fungi: Conti et al, 2021; Jiao & Yang, 2021; Lücking et al, 2020; Sipiczki et al, 2013, 2018). Genes different from the “standard barcode genes” can also be used for taxonomic and phylogenetic studies.…”

Near Chromosome-Level Genome Assembly and Annotation of <i>Rhodotorula babjevae</i> Strains Reveals High Intraspecific Divergence

“…The estimated genome divergence values through DDH, ANI and kr proved to be more sensitive for delineating Rhodotorula species. Phylogenetic placement based on the standard rDNA regions may not be enough to understand yeast diversity and species delineation as shown before [40,41]. These R. babjevae strains showed different behavior during enzymatic cell wall degradation for DNA purification within this study and when they were grown on xylose medium in another study [42].…”

MinION Sequencing of Yeast Mock Communities To Assess the Effect of Databases and ITS-LSU Markers on the Reliability of Metabarcoding Analysis

“…Whereas Illumina sequences for taxonomic metabarcoding normally cover only the ITS2 sequence, MinION can span the entire DNA region, including ITS and LSU D1/D2, thus increasing the amount of information (ca. 1,200 bp versus 400 bp) and therefore the taxonomic resolution ( 25 ). Mock communities were generated to assess the applicability of third-generation sequences for a comprehensive description of simulated microbiomes composed by uneven proportions of the different species, as happens in real situations.…”

“…The importance of highly curated and somehow focused reference databases has already been investigated at the level of all fungi and of pathogenic fungi in particular ( 39 ), showing that many sequences in public databases are too short or inaccurate or are derived from strains far away from the center of distribution of the species or from its type strain to be really good representatives ( 40 ). In addition to these problems, currently used markers in fungal taxonomy and barcoding cannot guarantee a full taxonomic resolution ( 25 , 41 ) as single gene protein-encoding markers ( 42 , 43 ), for which, however, universal anchoring positions are difficult to find, producing different levels of amplification in diverse taxa ( 44 ). Within these two rDNA markers, in this paper we showed that, in general, ITS-based reference databases work better than LSU ones.…”