A genome-informed higher rank classification of the biotechnologically important fungal subphylum Saccharomycotina

Abstract

The subphylum Saccharomycotina is a lineage in the fungal phylum Ascomycota that exhibits levels of genomic diversity similar to those of plants and animals. The Saccharomycotina consist of more than 1 200 known species currently divided into 16 families, one order, and one class. Species in this subphylum are ecologically and metabolically diverse and include important opportunistic human pathogens, as well as species important in biotechnological applications. Many traits of biotechnological interest are found in closely related species and often restricted to single phylogenetic clades. However, the biotechnological potential of most yeast species remains unexplored. Although the subphylum Saccharomycotina has much higher rates of genome sequence evolution than its sister subphylum, Pezizomycotina, it contains only one class compared to the 16 classes in Pezizomycotina. The third subphylum of Ascomycota, the Taphrinomycotina, consists of six classes and has approximately 10 times fewer species than the Saccharomycotina. These data indicate that the current classification of all these yeasts into a single class and a single order is an underappreciation of their diversity. Our previous genome-scale phylogenetic analyses showed that the Saccharomycotina contains 12 major and robustly supported phylogenetic clades; seven of these are current families (Lipomycetaceae, Trigonopsidaceae, Alloascoideaceae, Pichiaceae, Phaffomycetaceae, Saccharomycodaceae, and Saccharomycetaceae), one comprises two current families (Dipodascaceae and Trichomonascaceae), one represents the genus Sporopachydermia, and three represent lineages that differ in their translation of the CUG codon (CUG-Ala, CUG-Ser1, and CUG-Ser2). Using these analyses in combination with relative evolutionary divergence and genome content analyses, we propose an updated classification for the Saccharomycotina, including seven classes and 12 orders that can be diagnosed by genome content. This updated classification is consistent with the high levels of genomic diversity within this subphylum and is necessary to make the higher rank classification of the Saccharomycotina more comparable to that of other fungi, as well as to communicate efficiently on lineages that are not yet formally named.

Taxonomic novelties: New classes: Alloascoideomycetes M. Groenew., Hittinger, Opulente & A. Rokas, Dipodascomycetes M. Groenew., Hittinger, Opulente & A. Rokas, Lipomycetes M. Groenew., Hittinger, Opulente, A. Rokas, Pichiomycetes M. Groenew., Hittinger, Opulente & A. Rokas, Sporopachydermiomycetes M. Groenew., Hittinger, Opulente & A. Rokas, Trigonopsidomycetes M. Groenew., Hittinger, Opulente & A. Rokas.

New orders: Alloascoideomycetes: Alloascoideales M. Groenew., Hittinger, Opulente & A. Rokas; Dipodascomycetes: Dipodascales M. Groenew., Hittinger, Opulente & A. Rokas; Lipomycetes: Lipomycetales M. Groenew., Hittinger, Opulente & A. Rokas; Pichiomycetes: Alaninales M. Groenew., Hittinger, Opulente & A. Rokas, Pichiales M. Groenew., Hittinger, Opulente & A. Rokas, Serinales M. Groenew., Hittinger, Opulente & A. Rokas; Saccharomycetes: Phaffomycetales M. Groenew., Hittinger, Opulente & A. Rokas, Saccharomycodales M. Groenew., Hittinger, Opulente & A. Rokas; Sporopachydermiomycetes: Sporopachydermiales M. Groenew., Hittinger, Opulente & A. Rokas; Trigonopsidomycetes: Trigonopsidales M. Groenew., Hittinger, Opulente & A. Rokas.

New families: Alaninales: Pachysolenaceae M. Groenew., Hittinger, Opulente & A. Rokas; Pichiales: Pichiaceae M. Groenew., Hittinger, Opulente & A. Rokas; Sporopachydermiales: Sporopachydermiaceae M. Groenew., Hittinger, Opulente & A. Rokas.

Citation: Groenewald M, Hittinger CT, Bensch K, Opulente DA, Shen X-X, Li Y, Liu C, LaBella AL, Zhou X, Limtong S, Jindamorakot S, Gonçalves P, Robert V, Wolfe KH, Rosa CA, Boekhout T, Čadež N, Péter G, Sampaio JP, Lachance M-A, Yurkov AM, Daniel H-M, Takashima M, Boundy-Mills K, Libkind D, Aoki K, Sugita T, Rokas A (2023). A genome-informed higher rank classification of the biotechnologically important fungal subphylum Saccharomycotina. Studies in Mycology 105: 1–22. doi: 10.3114/sim.2023.105.01

This study is dedicated to the memory of Cletus P. Kurtzman (1938–2017), a pioneer of yeast taxonomy.

INTRODUCTION

Saccharomycotina yeasts are metabolically diverse, with great potential for use in biotechnological processes (Kurtzman et al. 2015). Furthermore, yeast species that are related often share biotechnology-relevant properties (Garay et al. 2016, 2017, Riley et al. 2016, Heistinger et al. 2022). For example, Riley et al. (2016) showed that some biotechnologically important traits, such as methylotrophy, lipogenesis, and xylose fermentation, as well as cellulose and hemicellulose hydrolysates utilisation, are mostly restricted to single phylogenetic clades. Currently, several Saccharomycotina species beyond Saccharomyces cerevisiae, so-called “non-conventional yeasts” (Spencer et al. 2002, Abbas 2003, Johnson 2013), serve as microbial cell factories for biotechnological applications, such as food and beverage fermentations (Bourdichon et al. 2012, Capece et al. 2018), biopharmaceutical molecules, biofuels and bioproducts from biomass, and compounds for the food, feed, pharmaceutical, and cosmetic industries (Groenewald et al. 2014, Hittinger et al. 2018, Spagnuolo et al. 2019, Gassler et al. 2020, Srinivasan & Smolke 2020, Solieri 2021, Abeln & Chuck 2021). Development of synthetic biology approaches, including CRISPR-Cas technologies, is expected to continue expanding the range of yeast species useful for biotechnology (Raschmanová et al. 2018, Cai et al. 2019, Patra et al. 2021). Historic classification and naming of yeast species were mostly restricted to analytical technologies and best practices and have resulted in classifications that do not always represent their evolutionary relationships. The revisions in this work are aimed to better align yeast species classification with evolutionary distance.

Traditionally, yeasts are separated from their filamentous counterparts based on a reproduction mode by budding or fission contrary to the production of hyphae (Flegel 1977). This delimitation has deep historical roots that date back to the early ages of mycology and to the separate study of yeasts and filamentous fungi (Lodder & Kreger-van Rij 1952). Although it is now well-established that many yeasts are dimorphic (i.e., also able to produce a filamentous stage), this fundamental separation is still widely used given its practicality. The definition that is best suited for yeasts is that they are ascomycetous or basidiomycetous fungi whose asexual growth predominantly results from budding or fission, with or without pseudohyphae and hyphae, and that they form sexual morphs that are not enclosed in fruiting bodies (Kurtzman et al. 2011a). The contemporary view acknowledges that the yeast stage has evolved multiple times across the fungal kingdom and, thus, should be viewed mainly as an ontological development and not as the diagnostic feature of any fungal group (Nagy et al. 2014). The so called “true yeasts”, by opposition to “yeast-like” (yeasts that form conspicuous mycelium often melanized and that are phylogenetically closer to filamentous taxa), are fungal lineages in the Ascomycota and Basidiomycota that adopted the yeast lifestyle, an assimilative stage predominantly based on unicellularity (Lodder & Kreger-van Rij 1952, Flegel 1977). A large portion of the known yeast taxa are found in the phylum Ascomycota and are divided into two subphyla, Saccharomycotina and Taphrinomycotina. The third subphylum in Ascomycota, Pezizomycotina, includes the ascomycetous filamentous fungi and some dimorphic yeast-like taxa (Spatafora et al. 2006, Kurtzman et al. 2011b, Nagy et al. 2014, James et al. 2020, Shen et al. 2020).

Most known ascomycetous yeast species belong to the subphylum Saccharomycotina (Shen et al. 2020), which is a lineage that exhibits levels of genetic diversity comparable to those of plants or animals (Dujon 2006, Kurtzman et al. 2011b, Shen et al. 2018, 2020, Li et al. 2021). Saccharomycotina yeasts inhabit a wide range of diverse habitats, such as plants, humans, animals, and terrestrial and aquatic ecosystems (Buzzini et al. 2017).

Many of the opportunistic human pathogens in Saccharomycotina are or were members of the genus Candida that grouped essentially all budding yeasts lacking sexual reproduction and other unique characteristics, most notably Candida albicans, C. tropicalis, Nakaseomyces glabratus, the C. parapsilosis species complex, and the emerging pathogen C. auris (de Hoog et al. 2020). Taxonomic revisions have placed Candida species in genera that better reflect their natural relationships, namely Clavispora lusitaniae (syn. Candida lusitaniae), Pichia kudriavzevii (syn. Candida krusei), Meyerozyma guilliermondii (syn. Candida guilliermondii), Nakaseomyces glabratus (syn. Candida glabrata), or will do so soon. To a lesser extent, species classified in the genera Saccharomyces, Kodamaea, Galactomyces, and Magnusiomyces have also been implicated as agents of fungemia, onychomycosis, and systemic disease, usually linked to immunocompromised patients, new-borns, and the elderly (Cooper 2011, de Hoog et al. 2020). Notably, human pathogenicity has evolved multiple times independently in the Saccharomycotina, including in N. glabratus, C. albicans, and C. auris (Gabaldon et al. 2016, Opulente et al. 2019, Rokas 2022).

An overview of the earlier classifications and trends can be found in the monographic series The Yeasts: A Taxonomic Study (Lodder & Kreger-van Rij 1952, Lodder 1970, Kreger-van Rij 1984, Kurtzman & Fell 1998, Kurtzman et al. 2011b). Before the advent of molecular sequencing the criteria for classifying yeasts were based on micro-morphological features of asexual and sexual reproduction and physiological / biochemical characteristics, with a strong emphasis placed on the ability to utilize different carbon and nitrogen compounds. These criteria divide the Saccharomycotina into two groups, yeasts which are known in its asexual (anamorphic) state and reproduce by forming asexual propagules (budding or fission), and yeasts which also show sexual (teleomorphic) morphs and can reproduce sexually by forming ascospores (Kurtzman et al. 2011b). Often, names of sexual morphs served as the basis for higher rank classifications. The Saccharomycotina, including most of the cultivatable genera of ascomycetous yeasts, currently contains a single class, Saccharomycetes, and a single order, Saccharomycetales (Kurtzman et al. 2011b). Thus, the inclusion of all budding yeasts in the Saccharomycetes and/or the Saccharomycetales was due to the broad definition that has been used for more than 140 years.

From the 1990’s onward, genetic sequence information became increasingly popular and immensely useful as a taxonomic criterion (Peterson & Kurtzman 1991, Kurtzman & Robnett 1997, 1998, 2013a, Kurtzman et al. 2003, Boekhout et al. 2021). Cletus Kurtzman gained prominence among yeast researchers by pioneering the use of molecular phylogenetic evidence to revise the taxonomy of the Saccharomycotina. He used sequences of the ribosomal RNA genes to group together genetically similar sexual and asexual species and genera of the Saccharomycotina yeasts, ensuring that clear affinities could be established (Kurtzman & Robnett 1997, 1998). During his career, he placed numerous Candida species and other asexual yeasts in existing sexual groups in the Saccharomycetales, making a great effort to address the problem of polyphyletic genera within this class.

Currently, molecular phylogenetic analyses based on multiple genes play a vital role in the taxonomy of yeasts included in the Saccharomycotina, not only in the delineation of novel species, but also in revising the classification of current asexual species (Daniel et al. 2014, Kurtzman 2016, Kurtzman et al. 2016, Yamazaki et al. 2020). In view of the requirement that a fungal species or higher rank taxon be assigned only a single valid name under the new International Code of Nomenclature for algae, fungi, and plants (Turland et al. 2018), the competing generic names of asexual/sexual forms (Kurtzman 2011, Kurtzman et al. 2011b), such as Blastobotrys/Trichomonascus, Botryozyma/Ascobotryozyma, Brettanomyces/Dekkera, Geotrichum/Dipodascus/Galactomyces, Kloeckera/Hanseniaspora, Myxozyma/Lipomyces, and Saprochaete/Magnusiomyces, as well as Candida species have been or are currently being revised to make genus membership consistent with phylogenetic affinities.

Analyses of multi-gene datasets that include representatives of all known genera and unresolved polyphyletic groups will facilitate stable generic and higher rank assignments and the placement of novel species, leading to an increase in the number of names and units to appropriately classify the given biological diversity. Currently, (novel) species and their classifications can be found on The Yeasts website (https://theyeasts.org/), the successor to The Yeasts: A Taxonomic Study book series. This website ensures that taxonomic changes are freely available as soon as possible after a classification is published.

The Saccharomycotina consists of more than 1 200 known species and 99 genera. They are currently divided into 15 recognized families (Alloascoideaceae, Ascoideaceae, Cephaloascaceae, Debaryomycetaceae, Dipodascaceae, Lipomycetaceae, Metschnikowiaceae, Phaffomycetaceae, Pichiaceae (invalid), Saccharomycetaceae, Saccharomycodaceae, Saccharomycopsidaceae, Trigonopsidaceae, Trichomonascaceae, and Wickerhamomycetaceae); Endomycetaceae is listed in Kurtzman et al. (2011b) as a speculative family within the Saccharomycotina. All these families are comprised in one order (Kurtzman et al. 2011b). Recent phylogenetic studies utilising genome-scale data have divided the Saccharomycotina into 12 major clades (Shen et al. 2017, 2018). These 12 clades have been referred to as the Lipomycetaceae, Trigonopsidaceae, Dipodascaceae/Trichomonascaceae, Alloascoideaceae, Sporopachydermia, CUG-Ala, Pichiaceae, CUG-Ser1, CUG-Ser2, Phaffomycetaceae, Saccharomycodaceae, and Saccharomycetaceae clades. Shen et al. (2018) also proposed the presence of additional higher order lineages within these major clades.

Recent phylogenomic analyses of ascomycetous fungi and yeasts, including comparisons of the evolutionary age and genomic divergence of taxonomic categories between the Saccharomycotina and other subphyla, suggest that current taxonomic circumscriptions do not fully account for the high level of evolutionary divergence in the Saccharomycotina (Shen et al. 2020, Li et al. 2021). Comparisons of genomic properties between the Saccharomycotina, consisting of one class, and its sister subphylum, Pezizomycotina, consisting of 16 classes, revealed that the two subphyla exhibited substantial divergence in several genomic properties (Shen et al. 2020). The Saccharomycotina have smaller genomes, lower numbers of protein-coding genes, lower numbers of DNA repair genes, lower GC contents, higher numbers of tRNA genes, and higher levels of genomic diversity and evolutionary rates than the Pezizomycotina (Shen et al. 2020). Still, the subphylum Saccharomycotina consists of only one class compared to the 16 classes in Pezizomycotina. The taxonomic placement of Ascomycete budding yeasts into a single class, Saccharomycetes, means that the last common ancestor of this class is much more ancient than the last common ancestors of any of the nine classes studied by Shen et al. (2020) in the subphylum Pezizomycotina. When comparing the fission yeasts (Taphrinomycotina) with the Saccharomycotina, the genome sizes of species belonging to these two subphyla are similar, but the number of species belonging to the Taphrinomycotina (~140 species) is nearly 10-fold lower than that of the Saccharomycotina, even though the Taphrinomycotina consists of six classes and six orders (Čadež et al. 2021).

In the study of Li et al. (2021), relative evolutionary divergence (RED) analyses showed that one of the most serious instances of under-classification within the Fungal kingdom concerns the Saccharomycotina. Simply put, there is little correspondence between the current poor taxonomic resolution and the observed very high evolutionary divergence within the Saccharomycotina. The present higher rank classification system is imbalanced on two levels, the under-classified class Saccharomycetes and the under-classified order Saccharomycetales.

Saccharomycotina species continue to be described in the Saccharomycetes and the Saccharomycetales. Given the high levels of evolutionary divergence detected in this group and considering the conclusions of previous phylogenomic studies (Shen et al. 2018, 2020, Li et al. 2021), it is clear that the higher taxonomic groups within the Saccharomycotina are due to be revisited and revised. A reclassification would be useful to communicate the evolutionary divergence more efficiently across biological disciplines. As the biotechnological industry currently uses only a small fraction of the more than 1 200 potentially exploitable Saccharomycotina species, reclassification of the higher ranks of the subphylum is expected to enhance the prospecting and discovery of novel biotechnological interesting yeasts and traits, as the focus of the search for traits of interest and novel species can then be more efficiently directed to clearly demarcated orders and classes.

Here we propose a new higher rank classification system that includes ten new orders and six new classes of the subphylum Saccharomycotina. Novel type families are described in the two cases for which it is essential for the description of the new orders and classes. The intent of our study is to focus on establishing higher level taxonomic boundaries that are evolutionarily consistent with the boundaries in other fungal lineages, as well as to set the groundwork for future studies that will address family- and genus-level circumscriptions. Using data from comprehensive genome-scale phylogenetic analyses, in combination with RED analyses, we propose the use of an updated classification of the Saccharomycotina that includes the following seven classes: Saccharomycetes, Pichiomycetes, Sporopachydermiomycetes, Alloascoideomycetes, Dipodascomycetes, Trigonopsidomycetes, and Lipomycetes. We similarly propose the following 12 orders: Saccharomycetales, Saccharomycodales, Phaffomycetales, Ascoideales, Serinales, Alaninales, Pichiales, Sporopachydermiales, Alloascoideales, Dipodascales, Trigonopsidales, and Lipomycetales.

MATERIALS AND METHODS

RED analyses

To evaluate whether fungal taxonomy is consistent with evolutionary genomic divergence, we calculated relative evolutionary divergence (RED) values using PhyloRank v. 0.0.37 (https://github.com/dparks1134/PhyloRank/) from the annotated tree as described previously (Li et al. 2021). Briefly, the NCBI taxonomy associated with every fungal genome was obtained from the NCBI Taxonomy FTP site on January 17, 2020. PhyloRank linearly interpolates the RED values of every internal node according to lineage-specific rates of evolution under the constraints of the root being defined as zero and the RED of all present taxa being defined as one (Parks et al. 2018). The RED intervals for each rank were defined as the median RED value ± 0.1 to serve as a guide for the normalization of taxonomic ranks from genus to phylum.

To evaluate whether the newly proposed taxonomic categories of Saccharomycotina were more consistent with evolutionary genomic divergence, we ran the RED analysis again with the proposed class and order names of Saccharomycotina.

Genome-scale phylogeny of the kingdom Fungi

The fungal phylogeny used in this study is derived from a recent phylogenomic study that used maximum likelihood analysis with a concatenation single-model (LG+G4) approach on a data matrix of 1 672 taxa (1 644 fungi and 28 outgroups) and 290 BUSCO genes (Li et al. 2021). See also Supplementary Table S1 for more information on the species used (Li et al. 2021).

Identification of class- and order-specific genes

To identify class- and order-specific genes, we used all annotated protein sequences in a previously published dataset of the genomes of 332 representative yeast species (Shen et al. 2018) to build orthologous groups of genes (OGs) using OrthoFinder v. 2.5.2 (Emms & Kelly 2019) with default parameters. For a given taxon of interest, we screened all 34 247 OGs and only retained those OGs with species occupancy ≥ 95 %, that is those OGs that were present in at least 95 % of species that were included in that taxon. In summary, we identified a total sum of 260 OGs and 325 OGs that were specifically present in the 7 proposed classes and 12 proposed orders, respectively. To further examine the specificity of those class- and order-specific OGs, we carried out TBLASTN searches of protein sequences of OGs against a custom database, consisting of the Nucleotide (nt) database downloaded from NCBI on December 11, 2021 and 1 644 fungal genomes from Li et al. (2021), with an e-value cut-off of 1e^-6. Lastly, we annotated the screened order-/class-specific OGs with Gene Ontology (GO) and Pfam domain families using the eggNOG-mapper v. 2.0 (Cantalapiedra et al. 2021).

Data availability Statement

All data associated with this study are provided in the main article and supplementary materials. The class- and order-specific orthologous groups (OGs) and their DNA and protein sequences are publicly available on the figshare repository (http://dx.doi.org/10.6084/m9.figshare.19374350).

RESULTS AND DISCUSSION

RED analyses

The availability of a genome-scale phylogeny for fungi provides a unique opportunity to evaluate if current taxonomy truly reflects the evolutionary relationships and rates of genome evolution of the Saccharomycotina. Previous taxon-rich phylogenomic analyses yielded a robustly supported phylogeny of the Saccharomycotina (Shen et al. 2018, 2020, Li et al. 2021). The new phylogeny divides the subphylum into 12 major clades, each of which is strongly supported by different approaches of maximum likelihood phylogenetic inference (e.g., concatenation and coalescence).

The RED (relative evolutionary divergence) approach was developed to revise taxonomic ranks in Bacteria and Archaea so that they reflect evolutionary divergence (Parks et al. 2018, Rinke et al. 2020). We recently applied this approach to fungi and found that fungal taxonomy is broadly aligned with both genome sequence divergence and divergence time (Li et al. 2021, Fig. 1A). However, we also determined that nearly 40 % (22 of 49 categories, including 1 order, 5 families, and 16 genera) of the under-classified categories in the taxonomy of fungi were within the Saccharomycotina (Fig. 1B), suggesting that the lack of taxonomic and evolutionary correspondence of the subphylum Saccharomycotina hinders the utility of taxonomy as a yardstick for comparative biology.

Fig. 1.

Relationship between relative evolutionary divergence (RED) and taxonomic rank between budding yeasts and other major fungal lineages. The RED intervals for each rank were defined as the median RED value ± 0.1 to serve as a guide for the normalization of taxonomic ranks from family to phylum. A. Data from Li et al. (2021) for the current order Saccharomycetales and class Saccharomycetes and other fungal lineages. B. Data for the newly proposed orders and classes. Note that when only a single lower rank is assigned to a higher rank (e.g. a single family is assigned to a single order), only the lower rank may be analyzed (e.g. Trigonopsidaceae, Dipodascales). Dipodascomycetes/Dipodascales (RED = 0.482) and Trigonopsidomycetes/Trigonopsidales/Trigonopsidaceae (RED = 0.439) still appear to be under-classified (i.e., had a much lower RED value than other ranks); whereas Serinales had a slightly lower RED value (RED = 0.633) compared to other newly proposed orders.

To improve the taxonomic classification of the Saccharomycotina, we propose new Saccharomycotina higher categories based on the current understanding of evolutionary relationships and rates of genome evolution of Saccharomycotina. In contrast to the single class and single order present in the previous Saccharomycotina higher rank classification, our proposed taxonomic revision divides the subphylum into seven classes and 12 orders.

Of the proposed seven classes and 12 orders examined, we found that the majority (five of seven classes, nine of 12 orders) of proposed categories fell within ±0.1 of the median RED value of taxa at that rank (a threshold previously used by Li et al. (2021)) as a guide to evaluate the degree to which the evolutionary divergences of taxonomic ranks across fungi were similar), suggesting the new categories better reflect the evolutionary divergence within the subphylum (Fig. 1b).

Phylogeny

Data from comprehensive phylogenomic analyses suggest that we can propose an updated classification of the Saccharomycotina. Specifically, we propose the following 7 classes: Alloascoideomycetes, Dipodascomycetes, Lipomycetes, Pichiomycetes, Saccharomycetes, Sporopachydermiomycetes, and Trigonopsidomycetes. Similarly, we propose the following 12 orders: Alaninales, Alloascoideales, Ascoideales, Dipodascales, Lipomycetales, Phaffomycetales, Pichiales, Saccharomycetales, Saccharomycodales, Serinales, Sporopachydermiales, and Trigonopsidales (Fig. 2, Supplementary Fig. S1, Supplementary Table S1).

Fig. 2.

A genome-scale phylogeny of Taphrinomycotina, Pezizomycotina, and Saccharomycotina. Maximum likelihood analysis with a concatenation single-model (LG+G4) approach on a data matrix containing 1 672 taxa (1 644 fungi and 28 outgroups) and 290 genes (Li et al. 2021). The orders and classes are indicated on the tree.

Class- and order-specific gene families

As described in the Materials and Methods, we identified a total of 260 OGs and 325 OGs that were specifically (≥ 98 % specificity) present in the seven proposed classes and 12 proposed orders respectively (Table 1). From the TBLASTN searches of protein sequences, we found that all 260 class-specific OGs and all 325 order-specific OGs had no significant blast hits elsewhere in the dataset (Supplementary Table S2). Results are presented in Table 1, Supplementary Tables S2, S3.

Table 1.

An overview of the most unique (98–100 % unique) class- and order-specific groups of orthologous genes (OGs) to use as diagnostic characters for the newly proposed orders and classes.

Rank	Order and Class names	nr of specific OGs	*OGs to use in description as diagnostic characters
Order	Saccharomycodales	18	OG0011566; OG0011567; OG0011580; OG0011587; OG0011592
Order	Saccharomycetales	2	OG0005235; OG0005246
Order	Phaffomycetales	2	OG0006529; OG0006543
Order	Ascoideales	20	OG0018641; OG0018642; OG0018656
Order	Serinales	2	OG0004743; OG0004801
Order	Alaninales	23	OG0016373
Order	Pichiales	2	OG0005494; OG0005374
Order	Sporopachydermiales	163	OG0028621; OG0028581; OG0028722; OG0028736
Order	Alloascoideales	33	OG0009556; OG0024318
Order	Dipodascales	3	OG0005588; OG0005810; OG0006132
Order	Trigonopsidales	2	OG0008190; OG0008482
Order	Lipomycetales	55	OG0010973; OG0011052; OG0008472; OG0009553
	Total number of OGs	325
Class	Saccharomycetes	2	OG0004556; OG000423
Class	Pichiomycetes	1	OG0000547
Class	Sporopachydermiomycetes	163	OG0028621; OG0028581; OG0028722; OG0028736
Class	Alloascoideomycetes	33	OG0009556; OG0024318
Class	Dipodascomycetes	4	OG0005588; OG0005810; OG0006132
Class	Trigonopsidomycetes	2	OG0008190; OG0008482
Class	Lipomycetes	55	OG0010973; OG0011052; OG0008472; OG0009553
	Total number of OGs	260

* See Tables S2 and S3 for more information on these proteins.

Updated higher rank taxonomy of the Saccharomycotina

Table 2 gives an overview of all the existing and newly proposed classes, orders, families, and genera of yeasts of the subphylum Saccharomycotina. Table 1, Supplementary Tables S2, S3 list all the class and order-specific protein families (OGs). See these tables for detailed information on these proteins.

Table 2.

Classes, orders, families, and genera of yeasts of the subphylum Saccharomycotina. Blue fonts denote newly proposed ranks. Red fonts denote taxa of uncertain assignment (incertae sedis = i.s.).

Class	Order	Family	Genus
Alloascoideomycetes	Alloascoideales	Alloascoideaceae	Alloascoidea
Dipodascomycetes	Dipodascales	Dipodascaceae	Dipodascus/Geotrichum
			Galactomyces/Geotrichum
			Magnusiomyces/Saprochaete
			Middelhovenomyces
		Trichomonascaceae	Crinitomyces
			Deakozyma
			Diddensiella
			Groenewaldozyma
			Limtongella
Saprochaete
			Spencermartinsiella
			Starmerella
						Sugiyamaella
			Trichomonascus/Blastobotys
			Wickerhamiella
			Zygoascus
	Dipodascales i.s.		Nadsonia/Schizoblastosporion
			Yarrowia
Lipomycetes	Lipomycetales	Lipomycetaceae	Babjevia
			Dipodascopsis
			Kockiozyma
			Lipomyces/Myxozyma
Pichiomycetes	Alaninales	Pachysolenaceae	Pachysolen
			Peterozyma
			Nakazawaea
	Pichiales	Pichiaceae	Allodekkera
			Ambrosiozyma
			Brettanomyces/Dekkera
Citeromyces
			Komagataella
			Kregervanrija
			Kuraishia
			Martiniozyma
			Ogataea
			Pichia
			Saturnispora
	Serinales	Cephaloascaceae	Cephaloascus
		Debaryomycetaceae	Aciculoconidium
			Candida
			Danielozyma
			Debaryomyces
			Diutina
			Hemisphericaspora
			Hyphopichia
			Kodamaea
			Kurtzmaniella
			Limtongozyma
			Lodderomyces
			Metahyphopichia
Pichiomycetes	Serinales	Debaryomycetaceae	Meyerozyma
			Millerozyma
			Nematodospora
			Priceomyces
			Scheffersomyces
			Spathaspora
			Suhomyces
			Schwanniomyces
			Teunomyces
			Wickerhamia
			Yamadazyma
		Metschnikowiaceae	Clavispora
			Metschnikowia
	Serinales i.s.		Babjeviella
Saccharomycetes	Ascoideales	Ascoideaceae	Ascoidea
		Saccharomycopsidaceae	Saccharomycopsis
	Phaffomycetales	Phaffomycetaceae	Cyberlindnera
			Phaffomyces
			Barnettozyma
			Starmera
		Wickerhamomycetaceae	Wickerhamomyces
	Saccharomycetales	Saccharomycetaceae	Cyniclomyces
			Eremothecium/Ashbya
			Grigorovia
			Hagleromyces
			Kazachstania
			Kluyveromyces
			Lachancea
			Nakaseomyces
			Naumovozyma
			Saccharomyces
			Savitreea
			Tetrapisispora
			Torulaspora
			Vanderwaltozyma
			Yueomyces
			Zygosaccharomyces
			Zygotorulaspora
	Saccharomycetales i.s.		Coccidiascus
			Endomyces
			Helicogonium
			Macrorhabdus
			Phialoascus
	Saccharomycodales	Saccharomycodaceae	Hanseniaspora/Kloeckera
			Saccharomycodes
Sporopachydermiomycetes	Sporopachydermiales	Sporopachydermiaceae	Sporopachydermia
Trigonopsidomycetes	Trigonopsidales	Trigonopsidaceae	Botryozyma/Ascobotryozyma
			Tortispora
			Trigonopsis

Taxonomy

Alloascoideomycetes M. Groenew., Hittinger, Opulente & A. Rokas, class. nov. MycoBank MB 847278.

Diagnosis: Class-specific protein families OG0009556 and OG0024318. Multilateral budding and formation of pseudohyphae and septate hyphae. Blastoconidia form on hyphae and may be sessile or arise from denticles. Sugars are not fermented. Phylogenetic analyses using DNA sequences encoding LSU rDNA, SSU rDNA, EF-1α, Rpb1, and Rpb2 (Kurtzman & Robnett 2013b).

Description: With the characteristics of Alloascoideaceae (Kurtzman & Robnett 2013b).

Type order: Alloascoideales M. Groenew., Hittinger, Opulente & A. Rokas ord. nov., this study.

Included order: Alloascoideales.

Notes: Although the number of species available for the orders Alloascoideales and Sporopachydermiales are limited, we propose the inclusion of these two orders in separate classes due to their long branches. See more information in the description of the class Sporopachydermiomycetes.

Alloascoideales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov. MycoBank MB 847279.

Diagnosis: Order-specific protein family OG0009556 and OG0024318. Phylogenetic analyses using DNA sequences encoding LSU rDNA, SSU rDNA, EF-1α, Rpb1, and Rpb2 (Kurtzman & Robnett 2013b). Multilateral budding and formation of pseudohyphae and septate hyphae. Blastoconidia form on hyphae and may be sessile or arise from denticles.

Description: With the characteristics of Alloascoideaceae (Kurtzman & Robnett 2013b).

Type family: Alloascoideaceae Kurtzman & Robnett, FEMS Yeast Res. 13: 429. 2013. MycoBank MB 802503.

Type genus: Alloascoidea Kurtzman & Robnett, FEMS Yeast Res. 13: 426. 2013. MycoBank MB 802504.

Type: Alloascoidea hylecoeti (L.R. Batra & Francke-Grosm.) Kurtzman & Robnett, FEMS Yeast Res. 13: 427. 2013. MycoBank MB 802506.

Family included in the Alloascoideales: Alloascoideaceae.

Genus included in the Alloascoideales based on the current study: Alloascoidea.

Notes: Although only one species is included in these analyses, considering the long branches observed by Kurtzman & Robnett (2013b) and Shen et al. (2018), we have decided to split off Alloascoidea from Sporopachydermia for inclusion in the new order Alloascoideales.

Dipodascomycetes M. Groenew., Hittinger, Opulente & A. Rokas, class. nov. MycoBank MB 847281.

Diagnosis: Class-specific protein families OG0005588, OG0005810, and OG0006132. This class contains dimorphic yeasts that can also produce arthroconidia. Phylogenetic analyses using DNA sequences encoding LSU rDNA, SSU rDNA, mtSSU rDNA, and EF-1α (Kurtzman et al. 2007).

Description: With the characteristics of Dipodascaceae (Engler & Gilg 1924).

Type order: Dipodascales M. Groenew., Hittinger, Opulente & A. Rokas ord. nov., this study.

Included order: Dipodascales.

Dipodascales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov. MycoBank MB 847282.

Diagnosis: Order-specific protein families OG0005588, OG0005810, and OG0006132. Phylogenetic analyses using DNA sequences encoding LSU rDNA, mtSSU rDNA, and Cox2 sequences (Kurtzman & Robnett 2007). Contains dimorphic yeasts that can also produce arthroconidia.

Description: With the characteristics of Dipodascaceae (Engler & Gilg 1924).

Type family: Dipodascaceae Engl. & E. Gilg, Syllabus der Pflanzenfamilien: 59. 1924. MycoBank MB 80709.

Type genus: Dipodascus Lagerh., Jahrb. Wiss. Bot. 24: 549. 1892. MycoBank MB 1632.

Type: Dipodascus albidus Lagerh., Jahrb. Wiss. Bot. 24: 549. 1892. MycoBank MB 246592.

Families included in the Dipodascales: Dipodascaceae, Trichomonascaceae.

Genera included in the Dipodascales based on the current and previous studies: Crinitomyces, Deakozyma, Diddensiella, Dipodascus (asexual morph Geotrichum), Galactomyces (asexual morph Geotrichum), Groenewaldozyma, Limtongella, Magnusiomyces (asexual morph Saprochaete), Middelhovenomyces, Nadsonia (asexual morph Schizoblastosporion), Spencermartinsiella, Starmerella, Sugiyamaella, Trichomonascus (asexual morph Blastobotrys), Wickerhamiella, Yarrowia, Zygoascus.

Notes: Genera that were not included in this study are Limtongella (Sakpuntoon et al. 2020b) and Crinitomyces (Sakpuntoon et al. 2022), but based on their phylogenetic placement, they are sister genera to Deakozyma and Dipodascus, respectively, which justifies the inclusion of both as genera of the Dipodascales. Based on our analyses, several genera that were previously included in the Saccharomycetales incertae sedis (Kurtzman et al. 2011b) are also included in this order, together with the genera of the families Dipodascaceae and Trichomonascaceae. Although the genus Trichomonascus was not included in the analyses, the asexual morph Blastobotrys was. Based on previous studies (Kurtzman & Robnett 2007, Kurtzman et al. 2011b) and due to the fact that Trichomonascus is the type genus of Trichomonascaceae, it is also included in the Dipodascales.

Our RED analyses showed low relative evolutionary divergence values, indicating possible continued under-classification for this group. However, when Yarrowia and Nadsonia were excluded from Dipodascales, this order’s RED value only shifted from 0.482 to 0.479, and it was still not located within the order boundaries. This led to the decision to include Yarrowia and Nadsonia in Dipodascales for now. However, description of more species in this group may eventually indicate that a split is necessary.

Biotechnological relevance: This order includes the biotechnologically important oleaginous yeast Yarrowia lipolytica, which is being considered for multiple industrial applications, such as a protein source for livestock feeding, a production host for organic acids or hydrophobic substances, for heterologous production of pharmaceutical and industrial proteins and enzymes, and for the production of bioproducts and biofuels (reviewed by Groenewald et al. 2014). Certain Blastobotrys species are used in biotechnology, including Blastobotrys mokoenaii, a species with a high secretion of xylanase, which was used industrially in bio-bleaching and bioprocessing of wood pulps (Techapun et al. 2003). Blastobotrys adeninivorans has major stress tolerant properties and an unusual metabolic flexibility that allows this species to use a wide range of carbon and nitrogen sources, which makes it a potentially important species for industrial purposes (Kunze et al. 2014). Starmerella species, especially those related to St. magnoliae, can produce erythritol (Yang et al. 1999) and D-mannitol (Song et al. 2002), which can be used as sweeteners. Starmerella bombicola is an excellent producer of sophorolipids (glycolipids), which are biodegradable and non-toxic biosurfactants, with a wide range of biotechnological applications in the food, health, and cosmetic industries (Takahashi et al. 2011, Qazi et al. 2022). Several species of Sugiyamaella are xylanase producers, and other species are reported to ferment D-xylose to ethanol, traits with biotechnological interest for the production of lignocellulosic ethanol (Sena et al. 2017).

Geotrichum klebahnii has several reported biotechnological characteristics and abilities, such as the rapid degradation of lignin from corn stover, good growth on hemicellulose from sugar-cane bagasse, and the secretion of propectinases with polygalacturonase activities. Geotrichum candidum is used to produce different varieties of cheeses and meat products (Ropars & Giraud 2022).

Clinical relevance: There have been reports of Galactomyces candidus, Magnusiomyces capitatus, and Yarrowia lipolytica causing human infections, and the first two species are increasingly found in the blood of immunocompromised patients (see overview in Kurtzman et al. 2011b, Groenewald et al. 2014, de Hoog et al. 2020).

Lipomycetes M. Groenew., Hittinger, Opulente & A. Rokas, class. nov. MycoBank MB 847285.

Diagnosis: Class-specific protein families OG0010973, OG0011052, OG0008472, and OG0009553. Phylogenetic analyses using either LSU rDNA or DNA sequences encoding SSU rDNA, LSU rDNA, mtSSU rDNA, and EF-1α (Kurtzman et al. 2007). Strains produce intracellular lipids and extracellular starch-like compounds, fermentation is absent, nitrate is not assimilated.

Description: With the characteristics of Lipomycetaceae (Novák & Zsolt 1961).

Type order: Lipomycetales M. Groenew., Hittinger, Opulente & A. Rokas ord. nov., this study.

Included order: Lipomycetales.

Notes: It is known that the higher levels of genomic diversity in Saccharomycotina stem from an acceleration of the evolutionary rate that occurred within the subphylum, after the divergence of Lipomycetaceae from the rest of the Saccharomycotina. This acceleration can also be seen in the genetic divergence that exists between the Lipomycetes and the rest of the Saccharomycotina classes presented here. See clustering presented in the phylogenomic tree in Shen et al. (2018), as well as Supplementary Fig. S1, which includes all groups of Saccharomycotina.

Lipomycetales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov. MycoBank MB 847286.

Diagnosis: Order-specific protein families OG0010973 OG0011052, OG0008472, OG0097159, and OG0009553. Phylogenetic analyses using LSU rDNA or DNA sequences encoding SSU rDNA, LSU rDNA, mtSSU rDNA, and EF-1α (Kurtzman et al. 2007). Ability to produce intracellular lipids and extracellular starch-like compounds. Fermentation is absent, nitrate is not assimilated.

Description: With the characteristics of Lipomycetaceae (Novák & Zsolt 1961).

Type family: Lipomycetaceae E.K. Novák & Zsolt, Acta Bot. Acad. Sci. Hung. 7: 97. 1961. MycoBank MB 80960.

Type genus: Lipomyces Lodder & Kreger, The Yeasts: A Taxonomic Study: 669. 1952. MycoBank MB 2888.

Type: Lipomyces starkeyi Lodder & Kreger, The Yeasts: A Taxonomic Study: 669. 1952. MycoBank MB 299844.

Family included in the Lipomycetales: Lipomycetaceae.

Genera included in the Lipomycetales based on this study and previous studies: Babjevia, Dipodascopsis, Kockiozyma, Lipomyces (asexual morph Myxozyma).

Notes: Although the genera Dipodascopsis, Babjevia, and Kockiozyma were not included in these analyses, based on previous analyses (Kurtzman et al. 2007, Yamazaki et al. 2020, Jindamorakot et al. 2012, Kurtzman et al. 2011a), it is clear that these genera belong to the family Lipomycetaceae and are therefore also included in the Lipomycetales.

Biotechnological relevance: There are many Lipomyces species with important biotechnological capabilities that enable them to play a role in the bioindustry. Lipomyces kononenkoae, L. lipofer, and L. starkeyi can hydrolyze dextran; produce d-12-desaturase, which is an enzyme that converts oleic acid into linoleic acid; produce dextranases; and show extracellular α-amylase activity, a key characteristic for the hydrolysis of industrial starch (Kurtzman et al. 2011b). Various Lipomyces species such as L. doorenjongii, L. kalimantanensis, L. lipofer, L. mesembrius, L. starkeyi, L. tetrasporus, L. tropicalis and L. yarrowii have been reported to be the oleaginous yeasts with cabable of accumulate high lipid, therefore, they have potential for microbial lipid production (Sapsirisuk et al. 2022, Poontawee et al. 2023).

Pichiomycetes M. Groenew., Hittinger, Opulente & A. Rokas, class. nov. MycoBank MB 847268.

Diagnosis: Class-specific protein family OG0000547.

Type order: Pichiales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov., this study.

Orders included in Pichiomycetes: Alaninales, Pichiales, Serinales.

Alaninales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov. MycoBank MB 847270.

Diagnosis: Order-specific protein family OG0016373. The CUG codon is translated into an alanine, instead of the nearly universal leucine. Phylogenetic analyses using DNA sequences encoding LSU rDNA, SSU rDNA, EF-1α, and mtSSU rDNA. Budding is multilateral and on a narrow base, ascospores are hemispherical or hat-shaped.

Type family: Pachysolenaceae M. Groenew., Hittinger, Opulente & A. Rokas, fam. nov., this study.

Pachysolenaceae M. Groenew., Hittinger, Opulente & A. Rokas, fam. nov. MycoBank MB 847271.

Diagnosis: Family-specific protein family OG0016373. The CUG codon is translated into an alanine, instead of the nearly universal leucine. Phylogenetic analyses using DNA sequences encoding LSU rDNA, SSU rDNA, EF-1α, and mtSSU rDNA (Kurtzman & Robnett 2010). Budding is multilateral and on a narrow base, ascospores are hemispherical or hat-shaped.

Type genus: Pachysolen Boidin & Adzet, Bull. Trimestriel Soc. Mycol. France 73: 340. 1958 [1957]. MycoBank MB 3676.

Type: Pachysolen tannophilus Boidin & Adzet, Bull. Trimestriel Soc. Mycol. France 73: 340. 1958 [1957]. MycoBank MB 302177.

Family included in the Alaninales: Pachysolenaceae.

Genera included in the Alaninales based on this study: Nakazawaea, Pachysolen, Peterozyma.

Notes: All genera assigned now in this order were previously included in the Saccharomycetales incertae sedis (Kurtzman et al. 2011b). For many years, the taxonomic placement of taxa in Nakazawaea, Pachysolen, and Peterozyma was unclear. Recent studies have shown that these genera form a group in which the CUG codon of species in these genera is translated into an alanine, instead of the nearly universal leucine. The CUG-Ala clade is monophyletic (Mühlhausen et al. 2016, Riley et al. 2016, Shen et al. 2016, 2017, 2018, Krassowski et al. 2018).

Biotechnological relevance: Nakazawaea peltata produces a β-glucosidase that can help, in combination with purified cellulase, in the degradation of cellulose; it is also an efficient producer of xylitol, a substance used increasingly as an artificial sweetener (Saha & Bothast 1996, 1999). Pachysolen tannophilus can convert crude glycerol feedstocks into ethanol (Liu et al. 2012), and it was also shown that strains from this species can switch from glucose to efficient xylose fermentation (Slininger et al. 2011).

Pichiales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov. MycoBank MB 847273.

Diagnosis: Order-specific protein families OG0005494 and OG0005374. Phylogenetic analyses using DNA sequences encoding LSU rDNA, SSU rDNA, EF-1α, and mtSSU rDNA (Daniel et al. 2014).

Type family: Pichiaceae M. Groenew., Hittinger, Opulente & A. Rokas, fam. nov., this study.

Pichiaceae M. Groenew., Hittinger, Opulente & A. Rokas, fam. nov. MycoBank MB 847274.

Synonym: ‘Pichiaceae’ Zender [as ‘Pichiacées’], Bull. Soc. Bot. Genève sér. 2, 17: 290. 1925. MycoBank MB 81172, nom. inval., Art. 32.1(c), see Art. 18.4 (Shenzhen).

Diagnosis: Family-specific protein families OG0005494 and OG0005374. Phylogenetic analyses using DNA sequences encoding LSU rDNA, SSU rDNA, EF-1α, and mtSSU rDNA (Daniel et al. 2014).

Type genus: Pichia E.C. Hansen, Centralbl. Bakteriol. Parasitenk., 2. Abth. 12: 538. 1904. MycoBank MB 4095.

Type: Pichia membranifaciens (E.C. Hansen) E.C. Hansen, Centralbl. Bakteriol. Parasitenk., 2. Abth. 12: 538. 1904. MycoBank MB 227217.

Family included in the Pichiales: Pichiaceae

Genera included in the Pichiales based on the current and previous studies: Allodekkera, Ambrosiozyma, Brettanomyces (teleomorph Dekkera), Citeromyces, Komagataella, Kregervanrija, Kuraishia, Martiniozyma, Ogataea, Pichia, Saturnispora.

Notes: Several genera that were previously included in the Saccharomycetales incertae sedis by Kurtzman et al. (2011b), together with the genera of the family Pichiaceae and the recently introduced genus Allodekkera (Jutakanoke et al. 2017), are now included in the Pichiales. Komagataella was previously included in the family Phaffomycetaceae (Kurtzman et al. 2011b), but genome-scale analyses support its reassignment to Pichiales (see also the description of Phaffomycetales).

Biotechnological relevance: Several species in the Ogataea clade were found to have characteristics that make them good candidates for biotechnological applications, with O. parapolymorpha as a model organism. Komagataella pastoris, Komagataella phaffii, Ogataea polymorpha, and Ogataea methanolica are the most frequently used yeasts for heterologous protein production (Ravin et al. 2013, Love et al. 2016, Riley et al. 2016, Liu et al. 2022). Strong regulatory promoters from Ogataea polymorpha, O. methanolica, and Candida boidinii were shown to be superior to endogenous promoters for recombinant protein expression in Komagataella pastoris (Mombeni et al. 2020, Vogl et al. 2020). Genome-scale metabolic models have have been constructed and metabolic engineering is being implemented in this clade, including protein production platform species in Komagataella and Ogataea (Patra et al. 2021, Vogl et al. 2020). Pichia kluyveri is common in natural fermentations of agricultural products, such as coffee beans and in beverages (Wang et al. 2020) and its potential importance in aroma production was also illustrated (Karaaliođlu & Yüceer 2021, Vicente et al. 2021). Candida arabinofermentans (Kurtzman & Dien 1998) and C. succiphila (Dien et al. 1996) can ferment L-arabinose into moderate concentrations of ethanol, and C. maris can perform the enantioselective reduction of 5-acetylfuro[2,3-c]pyridine to (S)-5-(1-hydroxyethyl)furo[2,3-c]-pyridine, which can be useful in the pharmaceutical industry (Kawano et al. 2003). Tiukova et al. (2019) suggested that Brettanomyces naardenensis has great biotechnological potential by providing evidence through functional genome annotations, for stress tolerance, and its ability to ferment D-xylose well. Brettanomyces bruxellensis is associated with fermentation processes in which it can also be considered a beneficial microorganism, specifically due to its unique nutrient metabolism and peculiar flavour production including in the production of specific beers (Steensels et al. 2015, Tiukova et al. 2019) and as a bioethanol producer (Steensels et al. 2015, da Silva et al. 2020).

Clinical relevance: Pichia kudriavzevii (asexual morph Candida krusei) is often isolated with humans and animals and was indicated to be the fifth most common cause of candidemia (see overview in Kurtzman et al. 2011b, de Hoog et al. 2020). Several Pichia species can grow at temperatures of 37 °C or higher, such as P. manshurica, P. inconspicua, and P. norvegensis, and these have been suggested to have the potential to infect humans and cause disease (Majoros et al. 2003, Kurtzman et al. 2011b).

Serinales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov. MycoBank MB 847269.

Diagnosis: Order-specific protein family OG0004743. The CUG codon is translated into a serine, instead of the nearly universal leucine.

Type family: Metschnikowiaceae Kamieński ex Doweld, Index Fungorum 33: 1. 2013. MycoBank MB 550284.

Type genus: Metschnikowia Kamieński, Trudy Imp. S.-Peterburgsk. Obshch. Estestvoisp., Vyp. 3, Otd. Bot. 30(1): 364. 1900. MycoBank MB 3147.

Type: Metschnikowia bicuspidata (Metschn.) Kamieński. Trudy Imp. S-Peterburgsk. Obshch. Estestvoisp., Vyp. 3, Otd. Bot. 30(1): 363. 1900. MycoBank MB 100690.

Families included in the Serinales: Cephaloascaceae, Debaryomycetaceae, Metschnikowiaceae.

Genera included in the Serinales based on the current and previous studies: Aciculoconidium, Babjeviella, Candida, Cephaloascus, Clavispora, Danielozyma, Debaryomyces, Diutina, Hemisphaericaspora, Hyphopichia, Kodamaea, Kurtzmaniella, Limtongozyma, Lodderomyces, Metahyphopichia, Metschnikowia, Meyerozyma, Millerozyma, Nematodospora, Priceomyces, Scheffersomyces, Schwanniomyces, Spathaspora, Suhomyces, Teunomyces, Wickerhamia, Yamadazyma.

Notes: This group is represented by species that translate their CUG codon into a serine, instead of the nearly universal leucine, and were included in the CUG-Ser1 clade (Shen et al. 2016, 2017, 2018, Riley et al. 2016, Krassowski et al. 2018). Based on these analyses, the medically important Candida species C. albicans, C. tropicalis, C. parapsilosis, C. orthopsilosis, and C. auris are included in the Serinales. Diutina, Hemisphaericaspora, Limtongozyma, Metahyphopichia, Nematodospora, and Schwanniomyces were not included in this study, but previous studies (Kurtzman & Suzuki 2010, Hui et al. 2014, Khunnamwong et al. 2015, Gouliamova et al. 2016, Boontham et al. 2020, Khunnamwong et al. 2022) placed these genera in Debaryomycetaceae or Metschnikowiaceae, two of the current families included in the Serinales.

Biotechnological relevance: This order includes the agro-industrially important genus Debaryomyces, particularly the species D. hansenii and its close relatives, which are known as extremophilic yeasts with great biotechnological potential (Fröhlich-Wyder 2003, Breuer & Harms 2006, see overview in Kurtzman et al. 2011b). Several species in the genera Debaryomyces and Metschnikowia are associated with fermented foods and are thus candidates for food applications (Bourdichon et al. 2012). Limtongozyma cylindracea has been of considerable interest for the ability to produce lipases that have numerous applications in the biotransformation of various molecules in the food, pharmaceutical, pesticide, and other industries (Matten et al. 2022). Several Scheffersomyces species, such as Sch. shehatae, Sch. insectosus, Sch. lignosus and Sch. stipitis are known for their ability to ferment <I>D-</I>xylose (reviewed by Ruchala & Sibirny 2021). Spathaspora passalidarum and Sp. arborariae are efficient D-xylose fermenters (Cadete & Rosa 2018). Additional species in this order have biotechnological potential. For example, Candida maltosa can assimilate n-alkane and therefore could decontaminate soil polluted by petroleum products (Schmitz et al. 2000). Candida mogii and C. tropicalis have been used in the conversion of <I>D</I>-xylose into xylitol (de Mancilha & Karim 2003, Kumar et al. 2022), and the opportunistic human pathogen C. tropicalis is also known for its abilities to degrade hydrocarbons, perform lipid biotransformations, and degrade polyphenols in wastewater (Ettayebi et al. 2003).

Clinical relevance: Candida albicans, C. tropicalis, C. auris, and C. parapsilosis with its close relatives C. metapsilosis and C. orthopsilosis are broadly known as opportunistic human pathogens that have been isolated from various clinical material (Lockhart et al. 2008, Miranda et al. 2009, Silva et al. 2009, de Hoog et al. 2020). Additional species that are seen as opportunists with strains isolated from immunocompromised patients are Candida dubliniensis (included in the Lodderomyces-Spathaspora clade), Clavispora lusitaniae, Meyerozyma guilliermondii, M. farinosa, and Kodamaea ohmeri (Gargeya et al. 1990, Page et al. 2006, Pfaller et al. 2006, Taj-Aldeen et al. 2006, Kurtzman et al. 2011b, Hong et al. 2018, de Hoog et al. 2020, Zhou et al. 2021).

Saccharomycetes G. Winter, Rabenh. Krypt.-Fl. Ed. 2, 1(1): 32, 68. 1881. emended by M. Groenew., Hittinger, Opulente & A. Rokas. MycoBank MB 90791.

Emended diagnosis: Class-specific protein families OG0004556 and OG0004235. See original taxonomic description in Winter (1881).

Type order: Saccharomycetales.

Included orders: Ascoideales, Phaffomycetales, Saccharomycetales, Saccharomycodales.

Ascoideales J.H. Schaffner, Ohio Naturalist 13 (4): 76 (1913), emended by M. Groenew., Hittinger, Opulente &, A. Rokas. MycoBank MB 90693.

Emended diagnosis: Order-specific protein families OG0018641, OG0018642, and OG0018656. CUG codons are preferentially translated into serine, instead of the nearly universal leucine. Cells are dimorphic, abundant development of true mycelium, often with blastoconidia. Asci are usually attached to hyphae. See original description in Schaffner (1913), which until now, has not been generally recognized by modern taxonomists.

Description: With the characteristics of Ascoideaceae (Engler 1892) and Saccharomycopsidaceae (von Arx & van der Walt 1987).

Type family: Ascoideaceae Engl., Syllabus: 26. 1892. MycoBank MB 816385.

Type genus: Ascoidea Bref., Untersuch. Gesammtgeb. Mykol. 9: 94. 1891. MycoBank MB 361.

Type: Ascoidea rubescens Bref., Untersuch. Gesammtgeb. Mykol. 9: 94. 1891. MycoBank MB 212818.

Families included in the Ascoideales: Ascoideaceae, Saccharomycopsidaceae.

Genera included in the Ascoideales based on this study: Ascoidea, Saccharomycopsis.

Notes: Ascoideales is represented by species that mostly translate their CUG codons into a serine, instead of the nearly universal leucine and were included in the CUG-Ser2 clade (Riley et al. 2016, Shen et al. 2016, 2017, 2018, Krassowski et al. 2018). One species, Ascoidea asiatica, translates CUG codons stochastically as either serine or leucine. All other studied species in the order, including A. rubescens and several Saccharomycopsis species, translate CUG only as serine (Riley et al. 2016, Krassowski et al. 2018, Mühlhausen et al. 2018, Junker et al. 2019). When the genome of only a single representative species was used in phylogenomic inference, the stability of the placement of the CUG-Ser2 clade in the phylogeny was influenced by the inclusion of an ortholog group consisting of orthologs of the S. cerevisiae <I>DPM1</I> gene (Shen et al. 2017). The inclusion of the genomes of three additional representative species from this clade eliminated the gene’s disproportionate influence and stabilized this clade’s placement within the Saccharomycotina phylogeny (Shen et al. 2018).

Biotechnological relevance: Several characteristics of Saccharomycopsis fibuligera can be further exploited, such as the ability to hydrolyse starch due to the presence of α-amylase and glucoamylase activities (Wickerham et al. 1944). The species can also actively accumulate trehalose when grown on starch (Chi et al. 2009). Many Saccharomycopsis species are capable of predation on fungi (Pimenta et al. 2008, Junker et al. 2018) and could be considered as biocontrol agents.

Phaffomycetales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov. MycoBank MB 847267.

Diagnosis: Order-specific protein families OG0006529 and OG0006543. Phylogenetic analyses using DNA sequences encoding SSU rDNA, LSU rDNA, EF-1α, Rpb1 and Rpb2 (Kurtzman & Robnett 2013a).

Description: With the characteristics of Phaffomycetaceae (Yamada et al. 1999) and Wickerhamomycetaceae (Kurtzman et al. 2008).

Type family: Phaffomycetaceae Y. Yamada et al., Biosci. Biotechnol. Biochem. 63: 831. 1999. MycoBank MB 82097.

Type genus: Phaffomyces Y. Yamada, Bull. Fac. Agric. Shizuoka Univ. 47: 30. 1997. MycoBank MB 27889.

Type: Phaffomyces opuntiae (Starmer et al.) Y. Yamada, Bull. Fac. Agric. Shizuoka Univ. 47: 30. 1997. MycoBank MB 445220.

Families included in the Phaffomycetales: Phaffomycetaceae, Wickerhamomycetaceae.

Genera included in the Phaffomycetales based on this study: Barnettozyma, Cyberlindnera, Phaffomyces, Starmera, Wickerhamomyces.

Notes: Komagataella was previously included in the family Phaffomycetaceae (Kurtzman et al. 2011b), but genome-scale analyses do not support its inclusion in the Phaffomycetales (this study, Shen et al. 2018), so Komagataella is included in the Pichiales (see the description of Pichales).

Biotechnological relevance: Species of the genus Wickerhamomyces have great biotechnological potential as many of them can grow under stressful environmental conditions, such as extremes of pH, low water activity, and anaerobic conditions (see overview in Kurtzman et al. 2011b, Chen et al. 2020, Sehnem et al. 2020). In particular, Wickerhamomyces anomalus is appreciated for its production of metabolites useful for biopreservation of of fruit and cereals and for other purposes (Walker 2010). This species has also shown promising potential to be used as a biocontrol agent against various fungal plant pathogens (Hashem et al. 2014, Oro et al. 2018, Limtong et al. 2020). Several species in the genera Cyberlindnera and Wickerhamomyces are associated with food fermentations and are thus candidates for applications involving human consumption (Bourdichon et al. 2012, Karaaliođlu & Yüceer 2021). Cyberlindnera jadinii (asexual morph Candida utilis) is a well-known fodder yeast and industrial producer of single-cell protein (SCP) (Salazar-López et al. 2022).

Saccharomycetales C. Luerssen, Grundz. Bot.: 156. 1877, emended by M. Groenew., Hittinger, Opulente & A. Rokas. MycoBank MB 816369.

Emended diagnosis: Order-specific protein families OG0005235 and OG0005246. See original description in Winter (1881).

Type family: Saccharomycetaceae C. Luerssen, Grundz. Bot.: 160. 1877. MycoBank MB 816370.

Type genus: Saccharomyces Meyen, Arch. Naturgesch. 4 (2): 100. 1838. MycoBank MB 542401.

Type: Saccharomyces cerevisiae (Desm.) Meyen, Arch. Naturgesch. 4 (2): 100. 1838. MycoBank MB 492348.

Family included in the Saccharomycetales: Saccharomycetaceae.

Genera included in the Saccharomycetales based on this and previous studies: Cyniclomyces, Eremothecium (asexual morph Ashbya), Grigorovia, Hagleromyces, Kazachstania, Kluyveromyces, Lachancea, Nakaseomyces, Naumovozyma, Saccharomyces, Savitreea, Tetrapisispora, Torulaspora, Vanderwaltozyma, Yueomyces, Zygosaccharomyces, Zygotorulaspora.

Notes: Cyniclomyces, Grigorovia, Hagleromyces, and Savitreea were not included in this study, but Kurtzman et al. (2011b) placed Cyniclomyces in the family Saccharomycetaceae, Sousa et al. (2014) showed that Hagleromyces is the sister genus to Cyniclomyces, Grigorovia is nested within Kazachstania (Gouliamova & Dimitrov 2020), and Savitreea is the sister genus to Lachancea (Sakpuntoon et al. 2020a); therefore, we have assigned these genera to Saccharomycetales.

Biotechnological relevance: This order includes the most commonly known biotechnologically important yeast genus Saccharomyces, with the species S. cerevisiae (Kurtzman et al. 2011b). The agro-industrially important yeast genera Kazachstania, Kluyveromyces, Torulaspora, and Zygosaccharomyces are also part of this order (Kurtzman et al. 2011b, Urien et al. 2019, Karim et al. 2020, Fernandes et al. 2021, Karaaliođlu & Yüceer 2021, Solieri 2021). Several species in Lachancea, Saccharomyces, Kazachstania, Kluyveromyces, Zygotorulaspora, and Zygosaccharomyces are associated with fermented foods, and are thus candidates for food applications (Bourdichon et al. 2012, Karaaliođlu & Yüceer 2021). Kluyveromyces marxianus, a thermotolerant species, shows high ethanol fermentation at high temperatures from sucrose, glucose, and xylose; therefore, it has high potential for industrial ethanol production in tropical areas (Limtong et al. 2007, Nonklang et al. 2008, Nitiyon et al. 2016). Torulaspora indica reveals high antagonistic activity againt fungal phathogens that cause plant diseases; thus, it has potential to be effective biocontrol agent (Konsue et al. 2020, Limtong et al. 2020). The metabolic functionalities of genera, including Kluyveromyces, Lachancea and Ogataea, have been mapped (Lertwattanasaku et al. 2015, Nanda et al. 2020, Kosaka et al. 2022, Lu et al. 2022).

Clinical relevance: Nakaseomyces glabratus, until recently known as Candida glabrata, is of great importance in clinical microbiology. This species is a major pathogen that accounts for an increasingly large proportion of nosocomial fungal infections, especially in immunocompromised patients (Fidel et al. 1999, de Hoog et al. 2020). In addition, N. bracarensis and N. nivariensis, are emerging human pathogens that are found in many patients in many countries. Although S. cerevisiae is classified as a safe organism, it has been found to colonize in the human body, in particular the human gut and genitourinary tracts (Belvoncikova et al. 2022), and several cases of S. cerevisiae fungemia have been reported, principally in immunocompromised people (see overview in Kurtzman et al. 2011b).

Saccharomycetales incertae sedis (i.s.)

Five genera, Coccidiascus, Endomyces, Helicogonium, Macrorhabdus, and Phialoascus were indicated as Saccharomycetales i.s. by Kurtzman et al. (2011b) as no living type cultures of these genera are available. For Endomyces, partial rDNA sequences available from two non-type strains of E. scopularum indicate that this genus is part of the Saccharomycetales; together with Helicogonium and Phialoascus, Endomyces has been assigned to the family Endomycetaceae (Kurtzman et al. 2011b). However, additional work is necessary to clarify the status of that family.

Saccharomycodales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov. MycoBank MB 847266.

Diagnosis: Order-specific protein families present OG0011566, OG0011567, OG0011580, OG0011587, and OG0011592. Phylogenetic analyses using DNA sequences encoding mtSSU rDNA, ITS, LSU rDNA, EF-1α, and Cox2 (Kurtzman et al. 2011b). Budding is bipolar, nitrate is not assimilated, the diazonium blue B reaction is negative, and coenzyme CoQ-6 is present.

Description: With the characteristics of Saccharomycodaceae (Kudryavtsev 1960).

Type family: Saccharomycodaceae Kudryavtsev, Die Systematik der Hefen: 270. 1960. MycoBank MB 81249.

Type genus: Saccharomycodes E.C. Hansen, Centralbl. Bakteriol. Parasitenk., 2. Abth. 12: 537. 1904. MycoBank MB 4814.

Type: Saccharomycodes ludwigii (E.C. Hansen) E.C. Hansen, Centralbl. Bakteriol. Parasitenk., 2. Abth. 12: 537. 1904. MycoBank MB 246456.

Family included in the Saccharomycodales: Saccharomycodaceae.

Genera included in the Saccharomycodales based on this and previous studies: Hanseniaspora/Kloeckera, Saccharomycodes.

Notes: Saccharomycodes was not included in this study, but Kurtzman et al. (2011b) placed Saccharomycodes in the family Saccharomycodaceae.

Biotechnological relevance: Several applications in biotechnological processes have been proposed for Saccharomycodes ludwigii, which can be investigated further. This includes production of (non-)alcoholic beverages (Romano et al. 1998, Methner et al. 2022, Vaštík et al. 2022), sulfidation of wine musts before fermentation (Gadzhiev et al. 1976), and production of vinegar due to its acetic acid resistance (Saeki 1990). Hanseniaspora species are common in many natural fermentations (Lleixà et al. 2016, Valera et al. 2021).

Sporopachydermiomycetes M. Groenew., Hittinger, Opulente & A. Rokas, class. nov. MycoBank MB 847275.

Diagnosis: Class-specific protein families OG0028621, OG0028581, OG0028722, and OG0028736. Phylogenetic analyses using DNA sequences encoding LSU rDNA, mtSSU rDNA, and Cox2 sequences (Kurtzman & Robnett 2007). Asexual reproduction is by multilateral budding on a narrow base. Glucose fermentation is absent or weak. Nitrate is not assimilated. myo-Inositol as sole carbon source is assimilated. Coenzyme Q-9 is formed.

Description: With the characteristics of Sporopachydermiales.

Type order: Sporopachydermiales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov., this study.

Included order: Sporopachydermiales.

Notes: Several phylogenetic studies found that the clades including Alloascoidea and Sporopachydermia are situated on long branches indicative of substantial evolutionary divergence, and their divergence times are similar or greater than that of the orders Pichiales and Alaninales (Lachance 2011, Kurtzman & Robnett 2013b, Shen et al. 2018). Although only two Sporopachydermia and one Alloascoidea species were included in the current study, we propose to describe two separate sister classes: Sporopachydermiomycetes to accommodate the genus Sporopachydermia, and Alloascoideomycetes to accommodate the genus Alloascoidea. See also the description of Alloascoideomycetes above. Although the number of validly described species available for these two classes is limited, Lachance (2011) listed a fair number of species of the genus Sporopachydermia that are awaiting formal descriptions.

Sporopachydermiales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov. MycoBank MB 847276.

Diagnosis: Order-specific protein families OG0028621, OG0028581, OG0028722, and OG0028736. Phylogenetic analyses using DNA sequences encoding LSU rDNA, mtSSU rDNA, and Cox2 sequences (Kurtzman & Robnett 2007). Asexual reproduction is by multilateral budding on a narrow base. Pseudohyphae and true hyphae are not formed. Glucose fermentation is absent or weak, nitrate is not assimilated, and myo-inositol is assimilated. Coenzyme Q-9 is formed.

Description: With the characteristics of Sporopachydermia (Rodrigues de Miranda 1978).

Type family: Sporopachydermiaceae M. Groenew., Hittinger, Opulente & A. Rokas, fam. nov., this study.

Sporopachydermiaceae M. Groenew., Hittinger, Opulente & A. Rokas, fam. nov. MycoBank MB 847277.

Diagnosis: Family-specific protein families OG0028621, OG0028581, OG0028722, and OG0028736. Phylogenetic analyses using DNA sequences encoding LSU rDNA, mtSSU rDNA, and Cox2 sequences (Kurtzman & Robnett 2007). Asexual reproduction is by multilateral budding on a narrow base, and pseudohyphae and true hyphae are not formed. Glucose fermentation is absent or weak, nitrate is not assimilated, and myo-inositol as sole carbon source is assimilated. Coenzyme Q-9 is formed.

Description: With the characteristics of Sporopachydermia (Rodrigues de Miranda 1978).

Type genus: Sporopachydermia Rodr. Mir., Antonie van Leeuwenhoek 44: 440. 1978. MycoBank MB 5165.

Type: Sporopachydermia lactativora Rodr. Mir., Antonie van Leeuwenhoek 44: 440. 1978. MycoBank MB 323856.

Family included in the Sporopachydermiales: Sporopachydermiaceae.

Genus included in the Sporopachydermiales based on the current study: Sporopachydermia.

Notes: Although only two species are included in these analyses considering the long branches observed by Kurtzman & Robnett (2013b) and Shen et al. (2018), we have decided to split off Sporopachydermia from Alloascoidea for inclusion in the new order Sporopachydermiales.

Trigonopsidomycetes M. Groenew., Hittinger, Opulente & A. Rokas, class. nov. MycoBank MB 847283.

Diagnosis: Class-specific protein families OG0008190 and OG0008482. Phylogenetic analyses using DNA sequences encoding LSU rDNA, mtSSU-rDNA, and Cox2 sequences (Kurtzman & Robnett 2007).

Description: With the characteristics of Trigonopsidaceae (Lachance & Kurtzman 2013).

Type order: Trigonopsidales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov., this study.

Included order: Trigonopsidales.

Trigonopsidales M. Groenew., Hittinger, Opulente & A. Rokas, ord. nov. MycoBank MB 847284.

Diagnosis: Order-specific protein families OG0008190 and OG0008482. Phylogenetic analyses using DNA sequences encoding LSU rDNA, mtSSU rDNA, and Cox2 sequences (Kurtzman & Robnett 2007, Lachance & Kurtzman 2013).

Description: With the characteristics of Trigonopsidaceae (Lachance & Kurtzman 2013).

Type family: Trigonopsidaceae Lachance & Kurtzman, Int. J. Syst. Evol. Microbiol. 63: 3113. 2013. MycoBank MB 803502.

Type genus: Trigonopsis Schachner, Z. Ges. Brauwesen 52: 137. 1929. MycoBank MB 10310.

Type: Trigonopsis variabilis Schachner, Z. Ges. Brauwesen 52: 137. 1929. MycoBank MB 281376.

Family included in the Trigonopsidales: Trigonopsidaceae.

Genera included in the Trigonopsidales based on this study and previous studies: Botryozyma (teleomorph Ascobotryozyma), Tortispora, Trigonopsis.

Notes: Trigonopsis, Tortispora, and Botryozyma were included in the family Trigonopsidaceae (Lachance & Kurtzman 2013). Unfortunately, since Trigonopsidales contains only a single family, it could not be included in the RED analyses. Considering the exceptional early divergence of the basal branching point marking this order, it is possible that additional data may eventually support the division of this new order. However, due to the lack of species, we have decided to combine the genera Trigonopsis, Tortispora, and Botryozyma into one order, while acknowledging that future descriptions of more species in this group may lead to the split of Tortispora and Botryozyma from Trigonopsis.

Biotechnological relevance: The ability of Trigonopsis variabilis to produce D-amino acid oxidases that catalyse the oxidative deamination of α-amino acids to α-keto acids and ammonia (see overview in Kurtzman et al. 2011b) is of great importance to the pharmaceutical industry. Other uses for this enzyme include synthesis of chiral intermediates to produce antihypertensive drugs (Patel 2001) and the conversion of cephalosporin C to 7-amino cephalosporanic acid, which is a key intermediate for cephem antibiotics (Dib & Nidetzky 2008).

CONCLUSIONS

Since the currently used classification of the higher taxonomic categories within the Saccharomycotina is inconsistent with evolutionary relationships and distances, we have proposed an updated class and order classification. Increasing the numbers of classes and orders in the Saccharomycotina from one to seven and twelve, respectively, makes the higher rank lineages more consistent with the high level of genomic diversity that exists within this subphylum and more comparable to the higher-level taxonomic ranks in use for other fungi, which is especially important for environmental sequencing and comparative studies. Without this proposal, the current class and order classifications would continue to be nearly worthless for macro-evolutionary comparisons because all budding yeasts of the subphylum Saccharomycotina are currently included in the Saccharomycetales and the Saccharomycetes. Taxonomy should ideally reflect evolutionary relationships among organisms, and genome-based phylogenies are suitable to guide taxonomy because they allow to consider both evolutionary relationships and differing rates of evolution. The newly proposed classification accounts for monophyly and divergence time, making the main higher taxonomic ranks more informative. The newly proposed higher-level taxonomic categories of Saccharomycotina provide a more natural classification and improve the taxonomic and phylogenetic precision of higher taxonomic levels. Another important aspect of the improved taxonomy is that it makes it clearer that the cause of being a pathogen might be different in different clades. Currently, (novel) species and their classifications can be found on The Yeasts website (https://theyeasts.org/), the successor to The Yeasts: A Taxonomic Study book series. This website ensures that taxonomic changes are freely available as soon as possible after the new classification is published.

Current taxonomies scale consistently with many other features of biological interest, such as magnitudes of divergence in morphology, physiology, behaviour, or ecology. Most of the newly proposed orders are tied to unique genetic characters (unique OGs, (multi)gene sequences, codon translation), although in some cases, distinct morphological and physiological characters can be used as supplemental features. For example, the formal description of the newly proposed orders references diagnostic characters for each as follows: Saccharomycetales (unique OGs), Saccharomycodales (OGs, barcoding, morphology, physiology), Phaffomycetales (OGs, barcoding), Alaninales (OGs, codon translation CUG-Ala, barcoding, morphology, physiology), Serinales (OGs, codon translation CUG-Ser), Ascoideales (OGs, codon translation CUG-Ser, morphology), Pichiales (OGs, barcoding), Lipomycetales (OGs, barcoding, physiology), Sporopachydermiales (OGs, barcoding, morphology, physiology), Alloascoideales (OGs, barcoding, morphology), Dipodascales (OGs, barcoding, morphology), and Trigonopsidales (OGs, barcoding, physiology). Although this current study only focused on the higher classifications of classes and orders, it is clear from the genome-scale phylogeny presented and earlier studies that the family Wickerhamomycetaceae is polyphyletic, and future genus and family-level reclassification within the Phaffomycetales is needed. Currently, two formally described families are included in the Dipodascales, Trichomonascaceae and Dipodascaceae, but the Yarrowia and Nadsonia species have not yet been placed in either of the two families; thus, the proposal of a third family within the Dipodascales is a realistic possibility. The three families of the Serinales do not accommodate the genus Babjeviella, which remains of uncertain family assignment. Its highly divergent phylogenetic position prevents us from proposing a fourth family at this time as new related taxa may influence the taxonomic view on this.

Since the current classification does not fully meet the expectations of ecologists and biodiversity researchers for the subphylum Saccharomycotina, we expect that this new classification proposed here will be helpful and understandable for taxonomists, ecologists, industrial, and biomedical users. This new Saccharomycotina classification offers great promise for classifying potentially uncultivable or novel species in different microbiomes, and it will be the key for reliable assignment of ecological and functional traits to taxa for further ecophysiological and biodiversity analyses.

To our knowledge, species that contain biotechnologically important traits are present in all the newly proposed orders, except Sporopachydermiales and Alloascoideales. However, only a handful of species are currently used as biotechnological production platforms, mostly those from the order Saccharomycetales. Since industry uses only a small fraction of the hundreds of species known in the Saccharomycotina linage, we expect that the introduction of this reclassification of the higher categories for this subphylum will make discovery and development of additional yeasts of biotechnological interest easier. For example, some biotechnologically important traits have been suggested to be restricted to single orders, such as the exceptional protein production systems of the Pichiales, and the new taxonomy can facilitate a more focused screening process. Alternatively, screens for traits of interest might choose a subset of taxonomically diverse yeasts.

Several opportunistic yeast species are primary agents of disease after a drop in patient immunity, while others cause secondary infections usually linked to patients suffering from immune deficiencies or are emerging human pathogens. Most of the major opportunistic pathogenic species belong to the order Serinales, but many emerging opportunistic pathogens also belong to the Saccharomycetales, the Pichiales, and the Dipodascales. Therefore, a broad range of orders within the Saccharomycotina include yeast species that have the potential to emerge in the future as opportunistic human pathogens, especially in light of the increasing numbers of immunocompromised patients (Stavrou et al. 2019, de Hoog et al. 2020). The new classification will help to limit further reclassification and better reflects substantial evolutionary divergence of species. For example, in the current classification, the distantly related N. glabratus and C. albicans were in the same class, but the new scheme assigns them to different classes. This highlights the fact that the two species evolved the ability to cause human infections independently and have different clinically relevant properties, such as resistance to azoles.

The phylogeny-informed taxonomic classification that we are advocating separates phylogenetically distant taxa to highlight their diverse characteristics and potentials, which are sometimes hidden in their genotypes. The improved possibilities to communicate these characteristics and potentials by appropriate group names (class, order, family) may unlock otherwise obscure research opportunities in the fields that include evolution, biogeography, and conservation biology. It will also benefit researchers involved in bioprospecting by identifying genes that are specific to higher rank taxa and suggesting which taxa should be studied as fungal models of particular physiologies or mined as a source of biological parts (i.e., genes, promoters, chemicals) for applications in biotechnology.

Among fungal lineages, prior to this study, the Saccharomycotina were clearly under-classified in the higher taxonomic lineages. Many families and genera, previously included in the Saccharomycetales incertae sedis (Kurtzman et al. 2011b), are now placed within the newly proposed orders as it became clear that they did not fit into the sole order Saccharomycetales. This study provides a model for application of genome-scale datasets to higher rank taxonomy. This revised, genome-enabled classification within Saccharomycotina provides a foundation and a roadmap upon which future taxonomic studies can be built. In the near future, more than 1 000 genomes cataloguing nearly every known species of Saccharomycotina (Y1000+ Project: http://y1000plus.org; Hittinger et al. 2015) will be publicly available. The availability of these genomes will further allow the depth and breadth of family and genus circumscriptions to be evaluated to improve the classification of these lineages by integrating genomic approaches into taxonomy, an integrative process being called taxogenomics (Libkind et al. 2020). We predict the approach being taken in budding yeasts will pave the way for other fungi and, ultimately, taxonomy across the entire tree of life will enter the genomic era.

Supplementary Material: https://studiesinmycology.org/

Fig. S1.

The genome-scale phylogeny of 1 644 species in the fungal kingdom (with 28 outgroups) from Li et al. (2021). The tree of 1 672 fungal species was reconstructed from the maximum likelihood concatenation analysis of 290 single-copy BUSCO genes under a single LG+G4 model. See also Fig. 2 and Table S3 for the current names as on 23 November 2022.

Table S1.

Updated taxonomy as on 23 November 2022, strain ID, and source information of 1 672 fungal and outgroup genomes from the genome-scale phylogeny of the fungal kingdom constructed by Li et al. (2021).

Table S2.

Overview of the class- and order-specific genes. Detailed results of the best OGs (98–100 % unique) used as diagnostic characters for the newly proposed orders and classes.

Table S3.

Overview of the class- and order-specific genes. Detailed results of all OGs specific to the proposed orders and classes.