Modern taxonomic classification is often based on phylogenetic analyses of a few molecular markers, although single-gene studies are still common. However, the use of one or few molecular markers can lead to inaccurate inferences of species history and errors in classification. Here, we leverage genome-scale molecular phylogenetics (phylogenomics) of species and populations to reconstruct evolutionary relationships in a dense dataset of 711 fungal genomes from the biomedically and technologically important genusAspergillus. To do so, we generated a novel set of 1,362 high-quality molecular markers specific forAspergillusand provide profile Hidden Markov Models for each, facilitating others to use these molecular markers. Examination of the resulting genome-scale phylogeny: (1) helped resolve ongoing taxonomic controversies and identified new ones; (2) revealed extensive strain misidentification, underscoring the importance of population-level sampling in species classification; and (3) identified novel lineages that may shed light on the early evolution of an important genus. These findings suggest that phylogenomics of species and populations can facilitate accurate taxonomic classifications and reconstructions of the tree of life.

Cryptic fungal pathogens pose significant identification and disease management challenges due to their morphological resemblance to known pathogenic species while harboring genetic and (often) infection-relevant trait differences. The cryptic fungal pathogenAspergillus latus, an allodiploid hybrid originating fromAspergillus spinulosporusand an unknown close relative ofAspergillus quadrilineatuswithin sectionNidulantes, remains poorly understood. The absence of accurate diagnostics forA. latushas led to misidentifications, hindering epidemiological studies and the design of effective treatment plans. We conducted an in-depth investigation of the genomes and phenotypes of 44 globally distributed isolates (41 clinical isolates and three type strains) fromAspergillussectionNidulantes. We found that 21 clinical isolates wereA. latus; notably, standard methods of pathogen identification misidentified allA. latusisolates. The remaining isolates were identified asA. spinulosporus(8),A. quadrilineatus(1), orA. nidulans(11). Phylogenomic analyses shed light on the origin ofA. latus, indicating one or two hybridization events gave rise to the species during the Miocene, approximately 15.4 to 8.8 million years ago. Characterizing theA. latuspangenome uncovered substantial genetic diversity within gene families and biosynthetic gene clusters. Transcriptomic analysis revealed that both parental genomes are actively expressed in nearly equal proportions and respond to environmental stimuli. Further investigation into infection-relevant chemical and physiological traits, including drug resistance profiles, growth under oxidative stress conditions, and secondary metabolite biosynthesis, highlight distinct phenotypic profiles of the hybridA. latuscompared to its parental and closely related species. Leveraging our comprehensive genomic and phenotypic analyses, we propose five genomic and phenotypic markers as diagnostics forA. latusspecies identification. These findings provide valuable insights into the evolutionary origin, genomic outcome, and phenotypic implications of hybridization in a cryptic fungal pathogen, thus enhancing our understanding of the underlying processes contributing to fungal pathogenesis. Furthermore, our study underscores the effectiveness of extensive genomic and phenotypic analyses as a promising approach for developing diagnostics applicable to future investigations of cryptic and emerging pathogens.

Fungi biosynthesize a diversity of secondary metabolites, small organic bioactive molecules that play diverse roles in fungal ecology. Fungal secondary metabolites are often encoded by physically clustered sets of genes known as biosynthetic gene clusters (BGCs). Fungi in the genusPenicilliumproduce diverse secondary metabolites that have been both useful (e.g., the antibiotic penicillin and the cholesterol-lowering drug mevastatin) and harmful (e.g., the mycotoxin patulin and the immunosuppressant gliotoxin) to human affairs. BGCs often also encode resistance genes that confer self-protection to the secondary metabolite-producing fungus. SomePenicilliumspecies, such asPenicillium lilacinoechinulatumandPenicillium decumbens, are known to produce gliotoxin, a secondary metabolite with known immunosuppressant activity; however, an evolutionary characterization of the BGC responsible for gliotoxin biosynthesis amongPenicilliumspecies is lacking. Here, we examine the conservation of genes involved in gliotoxin biosynthesis and resistance in 35Penicilliumgenomes from 23 species. We found homologous, less fragmented gliotoxin BGCs in 12 genomes, mostly fragmented remnants of the gliotoxin BGC in 21 genomes, whereas the remaining twoPenicilliumgenomes lacked the gliotoxin BGC altogether. In contrast, we observed broad conservation of homologs of resistance genes that reside outside the BGC acrossPenicilliumgenomes. Evolutionary rate analysis revealed that BGCs with higher numbers of genes evolve slower than BGCs with few genes. Even though the gliotoxin BGC is fragmented to varying degrees in nearly all genomes examined, ancestral state reconstruction suggests that the ancestor ofPenicilliumspecies possessed the gliotoxin BGC. Our analyses suggest that genes that are part of BGCs can be retained in genomes long after the loss of secondary metabolite biosynthesis.