It has been almost two decades since NGS was first invented, which marks the beginning of fungal high throughput sequencing. Continuously developing over the past decade, NGS not only acts as a gene sequencing tool but also acts as a microbial function detector. The main sequencing systems of NGS include the 454 sequencing system, the Illumina sequencing system, and the SOLiD system [8]. The NGS method that is now widely used was improved from the existing methods by Daryl M. Gohl’s group. A series of improvements, such as increased template concentrations, a reduced number of PCR cycles, and highly persistent polymerase and proofreading polymerase enzymes, have contributed to the increased accuracy of microbiome studies [9]. The error rate of NGS in sequence splicing is in the range of 0.1–15% [10]. For reducing the sequencing error rate of NGS, a new computational method named SequencErr was invented, which helped improve sequencing accuracy [11]. From a diagnostic perspective, identifying and removing bacterial, fungal, and viral genomic contamination are critical for NGS sequencing. This strategy was validated using DecontaMiner, which can be easily combined with standard NGS procedures to unmap contaminated sequences [12]. K-mer analysis not only allows clinical microbiological detection based on WGS and antimicrobial resistance (AMR) but also may be used for the genetic prediction of antibiotic sensitivity [13,14]. In addition, in terms of data processing, because the NGS platform is able to generate a large amount of sequence data, how to process these data poses a challenge to the bioinformatics analysis [8]. A user-friendly framework known as Orione provides a comprehensive computational pipeline that improves the user experience [15]. We have also compiled some bioinformatics pipelines for NGS data analyses from recent years and listed them in Table 1.

As a diagnostic technology, NGS should ideally be fast, convenient, accurate, broad-spectrum, and user-friendly, although the cost remains significant. There is still a long way to go to establish a novel, mature microbial detection platform for clinical and scientific testing. NGS has also been increasingly used to help diagnose fungal infections [19]. NGS provides the possibility of exploring fungal populations at the strain level. For example, by exploring Saccharomyces cerevisiae strains, researchers found that 13 variable genes could represent almost all of their phylogenetic information. And the same method used in yeasts can also be generalized to other individuals [20]. Studies have confirmed that operational taxonomic units (OTUs) over 500 bp in length are more helpful in studying fungal biomes. Based on the K-mer analysis workflow, a suitable k-mer size facilitates gene assembly for fungal strains [21]. Meanwhile, pool sequencing (iPool-Seq) is a large-scale insertion mutation screening method that improves toxicity factor screening in fungi [22]. We also collated the pros and cons of some NGS-based strategies for fungal pathogen detection in Table 2.

NGS is reported to be applied to some fungal infection-related diseases. First, invasive fungal diseases (IFDs) are local or systematic pathogen invasions responsible for high morbidity and mortality caused by various groups of fungal species, especially Candida spp. and Aspergillus spp. [28]. Kidd et al. [29] provided a detailed summary of the use of fungal molecular polymerase chain reaction (PCR) assays for investigating many kinds of IFDs, showing that PCR can be used to detect fungal species with reasonable specificity. Then, they suggested that NGS can be used for the discriminative analysis of fungal genetic diversity, including drug resistance identification and outbreak investigation. Second, mucormycoses are deadly IFDs, having high mortality, unavoidable disfiguring surgical treatment, and limited therapeutic options. Vincent M. Bruno et al. made conclusions on Mucormycoses and presented their objective insights. Unbiased NGS technology contributes to the related research of Mucorales, consisting of fungus–host interactions, diagnostic improvement, genome architecture, and others [30]. Moreover, Trichophyton rubrum is an opportunistic pathogen responsible for progressively expanding invasive diseases. NGS technology targeting microRNAs showed that the inactivated germinated T. rubrum microconidia co-cultured with human macrophages promoted the release of proinflammatory cytokines and changes in the regulation of microRNAs [31]. In addition, the group led by Ana Lúcia Fachin co-cultured T. rubrum CBS 118892 with human keratinocytes to assess the efficacy of terbinafine, a medicine used to treat dermatophyte infections [32]. Besides these examples, the first study to determine and evaluate nonmodel fungal genomes by modified NGS was conducted for paracoccidioidomycosis. After systematically improving the original Sanger sequence assembly, several genes critical to virulence and pathogenesis were studied more carefully.

The use of the NGS technology in fungal pathogen detection offers the following benefits: First, NGS technology is suitable for hostile culture and slow-growing microbial infections, such as fungi [33]. It will be a useful tool for low fungal loads [19]. Second, NGS provides more accurate identification of fungal species and is even more specific than other methods [4]. However, on account of technical reasons and objective errors, current NGS technologies are not entirely plausible to some extent. Regarding fungi, Leho Tedersoo [6] and his group provided their opinions and recommendations that many large and small issues are always hidden in the current NGS technology. In addition, both the repeatability of fungal sequencing data and the availability of public data are needed for mycobiome sequencing. Likewise, a research team conducted a study comparing the effect of the Respiratory Pathogen ID/AMR (RPIP) kit on a targeted NGS workflow. After comprehensive contrasts were implemented, they concluded that NGS workflows cannot replace the traditional culture and other techniques. The reason they cannot be replaced is the complexity of the bioinformatic analysis [34]. Moreover, different types of Internal Transcribed Spacer (ITS) primers may tendentiously lead to different types of fungi. For instance, ITS1-F, ITS1, ITS5, et al., have a bias toward basidiomycetes amplification, while ITS2, ITS3, ITS4, et al., are more eccentric to ascomycetes [35].

Currently, the applications of NGS, ranging from innovative diagnostic methods to routine clinical detection, are growing in leaps and bounds. Both the cost and the speed of detection have improved, allowing NGS to be used for routine microbial detection [36]. Especially since 2005, both the innovation and evolution of NGS technologies and the reduced costs of the required testing materials have promoted genomic testing [36]. To distinguish a broader spectrum of species, the repeatability, quantification results, and classification accuracy of NGS should be improved. NGS is still far from being implemented in the routine of diagnostic laboratories for fungal infections. Not only the prohibitive expense but also its popularity among clinicians lead to this result. With the advantages of reducing turnaround time, localized pathogen WGS should be sufficiently promoted [37]. The identification of fungi by NGS technology not only has its necessity but also still has room for future development.