Gut mycobiome maturation and its determinants during early childhood: a comparison of ITS2 amplicon and shotgun metagenomic sequencing approaches

Introduction: Microbial colonization of the gut in early life is important for the development of metabolism, immunity, and the brain. Fungi and bacteria both colonize the human infant gut. The relatively smaller contribution of fungi to the gut microbiome, as compared to bacteria, has posed technical challenges for the precise characterization of fungal communities (mycobiomes) and limited the ability to longitudinally examine mycobiome development.

Background: The aims of this study were to (1) characterize mycobiome maturation and identify clinical determinants of mycobiome compositional variation during the first 2 years of life and (2) compare two sequencing approaches (ITS2 amplicon and whole genome metagenomics) for characterizing mycobiome maturational features. Longitudinal fecal samples and associated clinical metadata were obtained from subjects enrolled as part of the MAGIC (Microbiome, Antibiotics and Growth Infant Cohort) study.

Results: Overall, fungal richness increased and mycobiome composition changed in a similar ordered pattern during the first 2 years of life utilizing either amplicon or metagenomic sequencing approaches. Less resolution of taxa to species and genera levels was observed for the metagenomic dataset. The predominant taxa identified by both sequencing approaches, Candida albicans, Saccharomyces/S. cerevisiae, and Malassezia restricta, each exhibited similar dynamics in abundances and prevalences over the first 2 years of life, irrespective of sequencing approach. Antibiotic exposure and breastfeeding status contributed to time-specific mycobiome compositional variation, results that were consistent for both types of sequence datasets. Candida albicans exhibited altered abundance dynamics in association with perinatal antibiotic exposure and birth mode for both sequencing approaches. Post hoc analyses suggested that the birth mode association could be driven by exposure to perinatal antibiotics in children delivered by cesarean section rather than by birth mode itself.

Discussion: In summary, amplicon and metagenomic sequencing approaches provide generally similar results with respect to mycobiome maturational dynamics and the contribution of clinical variables to variation. Differences in taxa identification by the two approaches likely due to sequence database differences, primer/genome sequence variation, and/or sequencing depth should be taken into consideration.

1 Introduction

Fungi are the dominant eukaryotes in the human gut (Parfrey et al., 2014) and are present as commensal organisms, transient colonizers, and pathogens. Although the majority of preclinical studies have focused on gut bacterial communities (microbiomes) for their mechanistic role in host physiology, fungal communities (mycobiomes) have also been shown to contribute, in particular to the function of the immune system and metabolism (Iliev et al., 2012; Everard et al., 2014). In addition, a recent study in mice showed that gut fungi are causally implicated in the assembly of bacterial microbiomes (van Tilburg Bernardes et al., 2020). Together, these results in animal models have highlighted the importance of including fungi along with bacteria in studies aimed at elucidating the role of microbiomes in human health and disease.

In humans, obesity, inflammatory bowel disease, and some childhood allergies have all been associated with alterations in the gut mycobiome (Ott et al., 2008; Mar Rodríguez et al., 2015; Sokol et al., 2017; Boutin et al., 2021). It remains, however, unclear if and how mycobiomes and their interactions with microbiomes are mechanistically involved in the development of these diseases. It is particularly important to understand microbiome/mycobiome effects on health and disease during infancy, the time of life when gut microbes are co-developing with host aspects of physiology, with the potential then to affect short- as well as long-term health outcomes. For example, mice treated with low-dose antibiotics during infancy develop obesity and obesity-related metabolic phenotypes that persist into adulthood, despite normalization of their microbiomes as juvenile animals (Cho et al., 2012; Cox et al., 2014), and some studies report human infants with disrupted gut microbiomes due to antibiotic use have an increased incidence of being overweight and developing atopy and asthma later in life (Ajslev et al., 2011; Murphy et al., 2014; Mitre et al., 2018).

The majority of microbiome studies in human infants have used amplicon-based sequencing to characterize fungi, targeting 18S, ITS1, or ITS2 ribosomal DNA sequences. Disadvantages of amplicon-based approaches include primer bias, limited taxonomic resolution and challenges in differentiating between closely related species (Schoch et al., 2012; Lücking et al., 2020). In contrast, whole metagenome sequencing, a more widely used approach in studies of bacterial microbiomes (Stewart et al., 2018; Vatanen et al., 2018; Olm et al., 2022) provides sequence data from fragments of genes from all kingdoms of microbes and the host in a sample. This facilitates direct comparative analyses, within and between studies, of many types of organisms using one sequencing run. Advantages of metagenomic sequencing (reviewed in Nilsson et al., 2019) include the avoidance of biases inherent to PCR and primer choice, as well as the ability to assemble new genomes, distinguish closely related species from each other, and identify the presence of gene pathways toward the inference of microbiome function. A disadvantage of the metagenomic approach for characterization of fungi is the challenge in identification of low-abundance fungi, which historically has been less of an issue with the PCR amplicon approach. As sequencing costs have decreased, datasets are improving with respect to the depth of fungal sequencing, raising the potential for metagenomic sequencing to be a reasonable approach for mycobiome research.

In this study, we sought to characterize early life gut mycobiomes and their determinants and compare the results obtained with amplicon sequencing to those with whole genome metagenomic sequencing. Based on advantages and disadvantages being present for both sequencing approaches, we thought it would be important to understand the extent to which conclusions agreed and differed between approaches, to inform future studies. Overall, we found similar results with respect to general mycobiome development and its determinants over the first years of life when using either of sequencing approach. Differences in results were also observed, for example, in the identification as well as abundance dynamics of some fungal taxa.

2 Materials and methods

2.1 Participant information and sample collection and storage

Subjects were recruited as part of the “Microbiome, Antibiotic and Growth Infant Cohort” (MAGIC) study approved by the Committee for the Protection of Human Subjects at Children’s Hospital of Philadelphia (CHOP) (IRB 15-012623). This study was approved by the Human Research Protection Program at the University of Minnesota (U of MN) (IRB ID STUDY00003037) for the use of de-identified subject clinical and demographic data from CHOP and the use and generation of microbial sequence data from subject fecal samples. Informed consent was obtained from study subjects. The study enrolled children born at Pennsylvania Hospital (Philadelphia, PA) receiving preventive healthcare in the CHOP Primary Care Network or participating private practices together with their biological mothers. Demographic distributions of study participants (race, ethnicity, and sex assigned at time of birth) matched the general trends seen in the participating sites. Enrolled subjects were required to be less than 120 h of age, weigh more than 2,000 g, greater than 36 weeks gestational age, and admitted for less than 120 h to the neonatal care unit. All mothers were over the age of 18 years and spoke English. Maternal and child demographic and clinical data were obtained from questionnaires and medical chart reviews. For antibiotic covariable analyses, children who received antibiotics, up to and including the relevant sampling time point after birth, were compared to those who had not had postnatal antibiotic exposure. For maternal perinatal (near the time of birth) antibiotic covariable analyses, children born to mothers who received antibiotics were compared to those whose mothers did not. Breastfeeding status was defined as “yes” if the subject was breastfeeding some, or all, of their feedings at the time of sample collection. Reporting of race in this study cohort was mandated by the US National Institutes of Health, consistent with the Inclusion of Women, Minorities, and Children policy. Child race was reported by the mother via survey that included seven race categories (American Indian/Alaska Native, Asian, Black or African American, Native Hawaiian or Other Pacific Islander, White, more than one race, or unknown) and two ethnicity categories (Hispanic and non-Hispanic). For exploratory analyses, non-White subjects were grouped together (Group 1) for comparison to White subjects (Group 2), due to low numbers of non-white subjects.

Fecal samples were collected from participants and health questionnaires were completed within a few days of birth, and every 3 months until the subject reached 24 months of age. All subjects did not provide samples for every time point, and some samples were obtained at non-standard time points. Supplementary Figure 1 summarizes the timing and number of samples obtained by individual subject. Pea-sized volumes of fresh fecal samples from diapers were collected by healthcare providers or study personnel during the birth hospitalization and by parents at home after discharge using OMNIgene-GUT fecal collection kits (DNAgenoTek Inc., Ottawa, Ontario, CA). Collection tubes contain transport media that stabilizes DNA, allowing transport of samples at ambient temperatures.

2.2 DNA extraction, library creation, and sequencing

DNA was extracted from fecal samples using the DNeasy PowerSoil kit (QIAGEN, Germantown, MD, United States) by the University of Minnesota Genomics Center (UMGC). For amplicon sequencing, fungal ITS2 amplicons were generated from extracted DNA via synthesis of sequencing primers alongside amplicons (Gohl et al., 2016) in a dual-indexing method using a forward primer targeting the 5.8S fungal DNA region (Heisel et al., 2022) and reverse primer RSeq targeting the 25S region (Heisel et al., 2015). Amplicon libraries were sequenced using the Illumina MiSeq platform and 2 × 300 V3 chemistry. Eleven mock samples containing synthetic ITS sequences (Palmer et al., 2018) and 11 negative control (water) samples were also sequenced to evaluate for the presence of extraneous fungal DNA contamination during the amplicon sequencing process. Synthetic ITS sequences, DNA sequences that do not match any known fungal (or microbial) sequences, were used as controls to determine the extent of DNA spill-over into adjacent sequencing lanes and to determine the amount, if any, of microbial DNA contamination as compared to negative water control samples. For metagenomic sequencing, DNA libraries were constructed from extracted DNA using the Illumina Nextera XT ¼ kit (Illumina, San Diego, CA, United States). Metagenomic libraries were sequenced on an Illumina NovaSeq, using the S4 flow cell with the 2 × 150 bp paired end V4 chemistry kit, and achieved a median sequencing depth of ~4.5 million total sequence reads per sample. Overall, fungal sequence reads were present in fecal samples at a median of one in every 2.2 million total metagenomic sequence reads, in general agreement with a prior study of infant mycobiomes [1 fungal read for every 1.4 million total reads (Auchtung et al., 2022)].

2.3 Sequence processing and taxonomy assignment

Resulting sequences in FASTQ format were stored and processed on hardware available through the Minnesota Supercomputing Institute. Raw FASTQ files from both the amplicon and shotgun metagenomic sequencing runs were initially processed individually through Shi7 (Al-Ghalith et al., 2018) version 0.9.9. Folders containing the raw sequence files were presented to shi7_learning.py, and the output from shi7_learning.py was used to inform and process the samples using shi7.py to generate quality controlled, trimmed, and multiplexed FASTA files (commands for both ITS and metagenomics: shi7_learning.py -i./fastq -o learnt, shi7.py -i./fastq -o./output --adaptor Nextera). For metagenomic sequences, 50 GB of memory and 24 processing cores were provided to shi7_learning.py to process four folders of approximately 1 TB of raw FASTQ files each; the time required for this shi7_learning step was several hours. For processing the metagenomic sequence files through shi7.py, a supercomputer node containing 900 GB of memory and 24 processing cores was used; processing generally required between 24 to 36 h to complete. Due to memory constraints, the metagenomic sequences were divided into four batches for processing by shi7, with four subsequent FASTA files generated. In contrast, the amplicon sequences required much less computational hardware and time for the generation of FASTA files and were able to be processed on an interactive terminal using desktop equivalent amounts of memory and processing power in a matter of minutes. Sequences remaining after processing via shi7 were then aligned to reference databases using BURST (Al-Ghalith and Knights, 2017) version 0.99. To align amplicon sequences, a reference database was generated from the ITS RefSeq collection compiled by the National Center for Biotechnology Information (NCBI) (Heisel et al., 2022) (command for ITS reads and BURST: burst12 -q combined_taxa.fna -a ITS_20180808.acx -r ITS_20180808.edx -b ITS_20180808.tax -o fungal_its.b6 -fr -t 24). The amplicon database contains the rRNA/ITS region from 8,902 fungal species. Sequences corresponding to the synthetic ITS controls were added to the amplicon database before use. To align metagenomic sequences, a database was generated from the RefSeq collection of reference fungal genomes compiled by the NCBI, accessed on February 17, 2020 (command for metagenomic reads and BURST: burst15 -q combined_taxa.fna -a Fungus02172020.acx -r Fungus02172020.edx -b Fungus02172020.tax -o fungal_wgs.b6 -fr -t 24). The fungal metagenomic database contains the full genome sequences from 301 fungal species. Manual curation of this database was required, as numerous genomes contained spurious contigs (i.e., sequences that were identified as either bacterial or human DNA by comparing suspicious contigs to the “nr” NCBI database using NCBI-BLAST), as previously noted by other researchers (Lind and Pollard, 2021). Contigs were identified as suspicious if they were found to align alongside non-fungal taxonomies when analyzing sequencing data using the -m ALLPATHS flag in BURST with a comprehensive bacterial and fungal RefSeq BURST database. For both sequencing approaches, alignment of sequences utilized BURST with a 95% identity cutoff flag with the forward/reverse complement flag activated. Processing the 9.4 GB amplicon FASTA file required significantly less computational power than the 200 GB metagenomic FASTA files through BURST (Amplicon: 60 GB of memory, 24 processing cores, approximately 3 h of time, Metagenomics: 990 GB of memory, 24 processing cores, with each of the 4 metagenomic sequencing files taking approximately 2.5 days of time). The resulting.b6 files were converted to reference and taxonomy tables using embalmulate (Al-Ghalith and Knights, 2017) with “GGtrim” activated for both sequence datasets (command for ITS reads and embalmulate: embalmulate fungal_its.b6 fungal_its.txt fungal_its_tax.txt “GGtrim,” command for metagenomic reads and embalmulate: embalmulate fungal_wgs.b6 fungal_wgs.txt fungal_wgs_tax.txt “GGtrim”).

2.4 Sequence analyses and data visualization

Taxonomy and metadata tables were imported into RStudio for analysis. The amplicon taxonomy table and the four metagenomic taxonomy tables, were pooled into a single metagenomic taxonomy table and subjected to quality control screening as follows. Sequence read counts were determined for each sample, and those samples residing in the lowest quartile of read counts were deemed to be potentially unrepresentative of the population due to their rarity and were discarded. The resulting sequence read cutoff value per sample was 39 for amplicon, and 69 for metagenomic sequencing approaches. Thus, the number of samples remaining (after quality control processes) for subsequent analyses was 773 for the amplicon sequence dataset and 756 for the metagenomic sequence dataset. To eliminate rare or spurious taxa, any taxon that was not present in at least five of the total number of samples for each sequencing approach was removed. Sequence data was normalized using either centered log ratio (CLR) transformation (for beta diversity determined by Euclidean distance) or median depth normalization (for taxa abundances). Taxonomic sequence data was transformed into proportionally transformed percentile abundances (for amplicon data) or actual abundances (for metagenomic data). Taxa counts were log transformed to convert results into normal distributions prior to statistical analyses. The diversity and specnumber functions in the vegan R package and lme function in the nlme R package were used to calculate and analyze alpha diversities (taxa count and Shannon diversity index were used). Taxa abundances over time were modeled using LOESS (LOcally Estimated Scatterplot Smoothing) regression and plotted with 95% confidence intervals in RStudio to visualize taxa abundance dynamics, and linear mixed-effects models were used to assess differences over time. Beta diversity analyses (PERMANOVAs) were performed using the pairwise Adonis function in the vegan/BiodiversityR packages in R. Data manipulations and visualizations were achieved with the following computational packages: ggplot2, ggbeeswarm, ggsignif, plyr, dplyr, tidyr, textshape, reshape2, and randomcoloR. To account for the possibility of false positive results due to multiple tests in the case of EnvFit analyses, p-values were adjusted using false discovery rate (FDR) correction with “FDR-corrected p” (q) < 0.25 considered statistically different. For longitudinal analyses of alpha diversity using the entire sample set, multiple sampling of subjects was corrected for by including subject ID as a random effect in statistical models. For longitudinal analyses of mycobiome beta diversity, samples were grouped into seven discrete “time clusters” (determined on the basis of similar numbers of samples in each grouping) according to the age of the subject at sample collection: cluster 1, 0–4 weeks; cluster 2, 5–13 weeks (2); cluster 3, 14–24 weeks; cluster 4, 25–47 weeks; cluster 5, 48–50 weeks; cluster 6, 51–63 weeks; cluster 7, 64–96 weeks. Only the first (“youngest”) sample from a given subject was included for each time cluster to control for multiple sampling (Stewart et al., 2018) (also noted in the text, tables, or figures where appropriate). Covariable contributions to mycobiome variation (beta diversity) for each time cluster were determined via the EnvFit function in RStudio. CLR-normalized fungal sequence data, with beta diversity distances determined by the Euclidean calculator, were used in EnvFit analyses for both the amplicon and metagenomic datasets, with statistical values generated from comparisons of the differences between variable centroids relative to the total variation in microbial beta diversity for each time cluster. We considered p-values <0.05, and FDR-corrected p (q)-values of <0.25 as statistically significant. Taxon-level associations with clinical variables that have q-values <0.25 are considered statistically significant in exploratory microbiome studies where further investigation and validation of results are necessary (Stewart et al., 2018; Stražar et al., 2021; Martin et al., 2022). In addition, for mycobiome comparisons between demographic and clinical groups (e.g., vaginal versus cesarean-section birth), potential covariables were added into the statistical models and, for the majority, did not contribute to significant associations between the primary variable with mycobiome variation, except where noted in the Results section.

3 Results

3.1 Study population characteristics

Overall, 773 fecal samples provided amplicon sequences, 756 samples provided metagenomic sequences, and 590 samples provided both types of sequence data for this study (after sequencing quality control measures were performed), which represented samples from 177 children (distribution of samples over time is shown in Supplementary Figure 1). Of these subjects, approximately 63% were delivered vaginally, 48% were female, 70% identified as White, and 54% were exposed to perinatal antibiotics. The mean gestational age at birth for the study cohort was 39.7 weeks (SD ± 1.1). Most of the children in this study (86%) had a birth weight between the 10th and 90th percentiles; 9% were below the 10th percentile and 6% were above the 90th percentile. Of the covariables considered in this study, only breastfeeding status and childhood antibiotic exposure significantly differed by subject age. The percentage of children engaged in breastfeeding significantly decreased over time from >95% during the first month of life to ~30% between 16–24 months of age, whereas those exposed to childhood (postnatal) antibiotics increased over time, from a low of <3% during the first month of life and reaching a plateau of ~50–60% in children over 6 months of age (Table 1).

Table 1

Table 1. Comparison of covariables by time of sample collection^a.

3.2 Identification of fecal fungi and comparison of amplicon to metagenomic sequencing approaches

Amplicon and metagenomic sequencing of fecal DNA identified 197 and 253 fungal taxa respectively; of these, 37 taxa were identified by both sequencing approaches. However, these common taxa comprised most of the abundant taxa with a combined abundance of 75% in the amplicon dataset and 60% in the metagenomic dataset. For either sequencing approach, the most abundant taxa were generally observed to also be the most prevalent for the cohort (Supplementary Table 2). Across both sequence datasets, the major fungal taxa in common were Candida albicans, Saccharomyces/Saccharomyces cerevisiae, Malassezia restricta, and Candida tropicalis. The amplicon dataset contained more unique Candida species whereas the metagenomic dataset contained more unique Aspergillus species. Of note, Candida parapsilosis, a common gut commensal and cause of infection during infancy, was identified as highly abundant and prevalent in this cohort, but only by the amplicon sequencing approach. In general, less resolution of taxa to species and genera levels was observed in the metagenomic dataset (96% of taxa for amplicon dataset vs. 89% of taxa for metagenomic dataset).

3.3 Gut mycobiomes are dynamic over the first 2 years of life

Fungal diversity features in the gut had similar dynamics over time in the children in this cohort, irrespective of sequencing dataset analyzed except for one feature, as discussed below. Fungal richness (taxa numbers) increased over time (Figure 1) in both datasets. Shannon diversity indices also increased over time, but only for the metagenomics dataset. The finding that Shannon indices had an opposite (decreasing) trend over time in the amplicon dataset suggests that there is a strong decline in evenness of fungal taxa abundances over time in this sequence dataset (Figure 1). In addition, fecal fungal community compositions (beta diversities) were significantly different for most of the pairwise comparisons between time clusters for both sequencing datasets (Supplementary Table 3). These mycobiome compositional dynamics are visualized in PCoA plots, with beta-diversity centroids moving from right to left in a generally ordered pattern as subject age increases and more heterogeneity in mycobiomes being observed among later time clusters, for both sequencing datasets (Figure 2). The finding that beta diversity did not differ between the time clusters corresponding to the oldest ages (clusters 6 and 7), for either sequencing dataset, may indicate that mycobiome composition is reaching a maturational plateau (Supplementary Table 3).

Figure 1

Figure 1. Dot plots of fungal alpha diversity over the first ~2 years of life using amplicon (left panels) and metagenomic (right panels) sequences. Alpha diversity measures: richness (top panels), Shannon index (bottom panels). Each dot represents an individual sample, and all samples are organized by time of collection as a continuous variable. Random sampling was used in linear statistical models to account for multiple samples per infant. Linear regression line of best fit, red; gray shading, 95% confidence intervals.

Figure 2

Figure 2. Principal coordinates analysis (PCoA) plots of fecal fungal taxonomic compositions (beta-diversity) by time cluster and sequencing approach. Ellipses show the standard error (SE) confidence limit set at 0.2, and crosshatches show the standard deviation (SD) with confidence limit set at 0.95. In the case of multiple samples from an infant in a time cluster, only the earliest sample was used in the analysis. Statistical comparisons of fungal beta diversity among time clusters and number of samples (n) within each time cluster for each sequencing dataset are found in Supplementary Table 4.

We also analyzed the relative abundances and prevalences of the prominent fungal taxa identified by both sequencing datasets. We observed that C. albicans, Saccharomyces/S. cerevisiae, and M. restricta had similar abundance and prevalence dynamics by either sequencing approach (Figure 3 and Supplementary Table 4). The abundance of C. albicans decreased and prevalence remained stable, Saccharomyces/S. cerevisiae abundance and prevalence increased, and M. restricta abundance and prevalence decreased (Figure 3). Although the abundance of C. tropicalis was generally stable over time by both sequencing approaches, its prevalence dynamics were not similar for amplicon and metagenomic datasets.

Figure 3

Figure 3. LOESS regression curves of relative abundances and prevalences of dominant fungal taxa during the first 2 years of life for amplicon and metagenomic sequence datasets. LOESS regression curves for individual taxa indicated by different colors; gray shading, 95% confidence intervals. Dashed line not detected but included for comparison.

3.4 Contributions of clinical and demographic variables to early life mycobiome compositional variation and comparison by sequencing approach

3.4.1 Alpha and beta diversity

As previously noted, mycobiome richness increased over the first 2 years of life, irrespective of the sequence dataset used. To understand if clinical and demographic variables were associated with differences in alpha diversity measures at specific times, we analyzed mycobiome richness and Shannon indices by clinical factor and for each individual time cluster (Supplementary Table 5). No variables consistently (across all time clusters) differed for either alpha diversity metric. Although some differences were observed for individual time clusters, these were not consistent across both sequence datasets. To examine the extent to which clinical and demographic variables explain the variance in mycobiome compositions (beta diversity) during early life, we performed EnvFit analysis by time cluster for both amplicon and metagenomic datasets. EnvFit reports the significance (p-value, comparing all variables within each time cluster) and amount of variance (r²) explained by each covariable in the statistical model.

Several variables were observed to contribute to beta diversity consistently across both sequencing approaches for some time clusters. For the earliest time (cluster 1), perinatal antibiotic exposure and birth mode explained the greatest amount of mycobiome variation, as compared to the other variables (note relative sizes of bars on Figure 4). Breastfeeding status was also a major contributor to variation at early times (cluster 2) for both sequencing datasets. At later times (clusters 4–6), childhood antibiotic exposure was the predominant consistent variable explaining mycobiome variation, whereas for the latest time (cluster 7), the greatest amount of mycobiome variation was explained by breastfeeding status and race.

Figure 4

Figure 4. Bar plots of explained variance of six potential mycobiome covariables modeled by EnvFit for amplicon and metagenomic sequence data. Horizontal bars represent the amount of variance (r²) explained by each covariable in the model. Covariables that contributed significant differences in mycobiome composition (p-value <0.05 with q-value of <0.25) are represented in bold font. Asterisk denotes statistical significance within a time point. For amplicon dataset: time cluster 3, infant race (p = 0.017, q = 0.05) and infant sex (p = 0.015, q = 0.05); time cluster 4, childhood antibiotics (p = 0.005, q = 0.03); time cluster 6, breastfeeding status (p = 0.027, q = 0.16). For metagenomic dataset: time cluster 2, breastfeeding status (p = 0.039, q = 0.23); time cluster 5, childhood antibiotics (p = 0.001, q = 0.006). NA, not able to perform statistical analysis due to inadequate number of subjects within one of the comparison groups (e.g., not breastfeeding and exposed to childhood antibiotics for time cluster 1).

3.4.2 Fungal taxa abundances

Several clinical variables were associated with differences in the abundance dynamics of the predominant fungal taxa (featured on Figure 3) in the feces of the study cohort during the first 2 years of life. Variables that showed significant associations with taxa abundance differences over time for both sequencing approaches were childhood antibiotic exposure, breastfeeding status, perinatal antibiotic exposure and birth mode (Supplementary Table 6). Childhood antibiotic exposure was associated with differences for Saccharomyces/S. cerevisiae abundance dynamics, with exposure being associated with generally higher abundances of Saccharomyces/S. cerevisiae (positive beta-coefficient, Supplementary Table 6), in particular at <6 months of age (Supplementary Figure 2). Breastfeeding was associated with differences in C. albicans (higher, positive beta-coefficient) as well as Saccharomyces/S. cerevisiae (lower, negative beta-coefficient) abundances over time (Supplementary Table 6), but the dynamics differences were not completely consistent between the two sequencing approaches (Supplementary Figure 2). Exposure to perinatal antibiotics was associated with higher C. albicans abundances throughout the time course (positive beta-coefficient, Supplementary Table 6 and Supplementary Figure 2), however, this association did not remain significant when birth mode was added into the statistical model. Similarly, C. albicans abundances tended to be higher over time in children delivered by cesarean section (positive beta-coefficient, Supplementary Table 6 and Supplementary Figure 2), but the difference did not remain significant when the perinatal antibiotic exposure variable was included in the statistical model. The interaction between birth mode and perinatal antibiotics aligns with the fact that antibiotics were administered to all mothers at the time of delivery of their infants by cesarean section. To attempt to separate the effects of perinatal antibiotics from those of birth mode, C. albicans abundances over time were compared by birth mode in the subset of infants exposed to perinatal antibiotics. In this comparison, no difference in C. albicans abundance dynamics were observed between birth mode groups (linear model, p > 0.05), which suggests that perinatal antibiotics may be the major driver of C. albicans abundance variation observed in children delivering by cesarean section in this cohort.

4 Discussion

Metagenomic sequencing, as compared to amplicon-based sequencing, has not been widely used to characterize fungal communities in humans, and only rarely in infant cohorts. The reasons for this have included the lower amounts of fungi (compared to bacteria) in human microbiomes, the higher costs for increased depth of sequencing to capture fungi, and less coverage of fungi in metagenomic sequence databases. In a recent study comparing amplicon (ITS2) and whole genome metagenomic sequencing approaches for the study of gut mycobiomes in healthy adults, general concordance was observed with respect to the identification of fungi, with Candida, Saccharomyces, and Malassezia species being among the most abundant fungal taxa (Nash et al., 2017). The detailed analysis of mycobiome longitudinal dynamics, and clinical determinants, however, was not explored and compared between approaches. With continued technological improvements and lower costs facilitating increased depth of sequencing, whole genome metagenomic sequencing may, in some cases, be a feasible strategy allowing the comparison of multiple microbial kingdoms, simultaneously with the genome of the host, using a single sequencing approach. The resultant understanding of cross-kingdom interactions would ultimately inform a more complete mechanistic picture of how microbiomes are involved in the development of human health and disease.

Prior studies of mycobiome development in infants have primarily utilized an amplicon sequencing approach, targeting ITS1 and ITS2 regions of the fungal rDNA locus, to characterize fungal communities (Heisel et al., 2015; Boutin et al., 2021; Auchtung et al., 2022; Gutierrez et al., 2023). Recently, two studies reported the use of metagenomic sequencing to characterize features of infant gut mycobiomes (Lind and Pollard, 2021; Auchtung et al., 2022). Altogether, studies utilizing either sequencing approach have found that Candida and Saccharomyces species were the major fungal taxa identified in the feces of young children. Consistent with these prior reports, we also found that amplicon and metagenomic sequencing approaches identified Candida and Saccharomyces species as predominant fecal taxa during early life. In the comparison of the two sequencing approaches for their ability to identify fungi, we found that ~20% of the total identified sequences were shared between the sequencing approaches. The taxa that were identified by both methods, however, comprised 60% of the fungal sequences present in the metagenomic dataset and 72% of the sequences in the amplicon dataset. Thus, although differences exist with respect to the identification of less common fungi, amplicon and metagenomic sequencing approaches have the potential to provide consistent results with respect to predominant mycobiome features and their variation.

Overall, we found general consistency between amplicon and metagenomic sequencing approaches with respect to how gut mycobiome features develop over the first years of life. Fungal richness increased over time and variation in fungal compositions (beta diversity) occurred with a similar ordered pattern for both sequencing approaches. The findings that mycobiome diversity and composition are dynamic and change during early life is a theme echoed by several previous studies utilizing either amplicon or metagenomic sequencing approaches (Fujimura et al., 2016; Schei et al., 2017; Wampach et al., 2017; Amenyogbe et al., 2021; Boutin et al., 2021; Auchtung et al., 2022; Gutierrez et al., 2023). In addition, we observed similar abundance and prevalence trends for the major fungal taxa C. albicans, Saccharomyces/S. cerevisiae, and M. restricta during early life, irrespective of sequencing approach, and are consistent with abundance dynamics reported by others (Lind and Pollard, 2021; Auchtung et al., 2022; Gutierrez et al., 2023).

Previous studies have observed that antibiotic exposure is a major contributor to gut bacterial compositional variation (Dethlefsen et al., 2008; Antonopoulos et al., 2009; Robinson and Young, 2010; Dethlefsen and Relman, 2011). In the current study, perinatal and childhood antibiotic exposure were major clinical variables associated with mycobiome feature variation over the first years of life. The result that antibacterial antibiotics have a trans-kingdom effect on fungal communities is consistent with the findings of other studies of gut mycobiomes (Heisel et al., 2022; Ventin-Holmberg et al., 2022). In the Ventin-Holmberg et al. (2022) study, antibiotic exposure in infants was associated with lower bacterial diversity and higher fungal diversity and abundances of several fungi, including C. albicans. This supports the idea that antibacterial antibiotics modulate mycobiomes via disruption of the normal balance of the bacterial community. Perinatal antibiotic exposure contributed more to mycobiome variation (beta diversity) during early times after birth, with generally higher abundances of C. albicans not only at early times but also over the entire time course. This result appears to be independent of birth mode. We did not have access to information about the reasons for perinatal antibiotic use, thus, we cannot rule-out the possibility that additional covariables contributed to this result. In contrast to perinatal antibiotics, childhood antibiotic exposure explained more of the compositional variation in mycobiomes at older ages and was associated with generally higher abundances of Saccharomyces/S. cerevisiae over time, although differences in Saccharomyces/S. cerevisiae abundances were not completely consistent across sequencing approaches. The finding that increased S. cerevisiae abundances may be associated with antibacterial antibiotic exposure is consistent with a recent report that broad spectrum antibiotic administration to mice enhances their ability to be colonized with Saccharomyces species in the gut (Hedin et al., 2022).

Our study, along with those of other investigators (Boutin et al., 2021; Auchtung et al., 2022), found that breastmilk feeding is associated with mycobiome composition during early childhood. Breastmilk is a complex, dynamic fluid that contains thousands of distinct components including nutrients and non-nutritive “bioactive” components [human milk oligosaccharides (HMOs), immune cells, bacteria and fungi] that promote healthy immune, metabolic and organ development (Bardanzellu et al., 2020). Milk nutrients may affect mycobiome compositions directly via the provision of nutrients that favor specific fungi over others. Bioactive components like HMOs, which promote the increased abundances of probiotic bacteria (Lactobacillus, Bifidobacteria) in the infant gut, and milk bacterial microbiomes themselves, could affect mycobiome compositions indirectly via interkingdom (bacterial-fungal) or immune cell-mediated interactions (van Tilburg Bernardes et al., 2020). Breastmilk fungi, the presence of which has been reported by several investigators (Heisel et al., 2019; Boix-Amoros et al., 2017; Moossavi et al., 2020), is a potential direct source of fungi for developing infant gut mycobiomes. Overall, much work remains to be done to determine how breastmilk components are inter-related with each other and with the infant gut in modulating early life microbiome development.

We acknowledge that our comparison of sequencing approaches is limited by the sequencing depths chosen, particularly for metagenomic sequencing. While the depth of metagenomic sequencing in the current study is similar to that reported in the study of infant gut microbiomes by Auchtung et al. (2022), approximately 1–2 gigabases/sample, it is ~10-fold less than that reported by the Human Microbiome Project study of adult gut microbiomes (Nash et al., 2017). With respect to PCR-based amplicon sequencing, this approach did not detect some taxa that the metagenomics approach did in our study. This result may be due, in part, to sequence mismatches between the ITS2 primers and the sequences of some fungal taxa, resulting in sub-optimal amplification. Even though ITS2 sequence databases include many more fungal taxa than metagenomic databases, it is also possible that they contain gaps in coverage. Finally, we did not include a host depletion step (either pre-sequencing or during the bioinformatics process) for the metagenomics process, thus, there is a chance that a small number of human reads identical to a reference fungal sequence could align to the fungal reference database, reducing the accuracy of our final metagenomic dataset. On the other hand, inclusion of a host sequence depletion step could also result in the rejection of true fungal sequences that align with regions of conservation in the host reference sequence database. Despite these considerations, this study has several strengths. A large infant cohort with a high number of longitudinal samples is represented in the analysis. In addition, we obtained high quality antibiotic exposure data for the entire cohort (duration and timing of exposure). This was due to collaborations with the Children’s Hospital of Philadelphia Health Network, the nation’s largest integrated pediatric health care system that provides electronic medical records connecting hospital and clinic data for children from birth onward and the Department of Biomedical and Health Informatics, which tracks office- and telephone-based prescriptions for study subjects and shares a common electronic health record with the CHOP Health Network. The quality of longitudinal antibiotic exposure data in this study may have allowed us to observe associations between antibiotics and early life mycobiome features that other studies have not.

In summary, our study supports the potential for metagenomic sequencing to be a useful approach for the characterization of gut mycobiomes and the identification of factors that contribute to fungal community variation. Higher sequencing depths and further improvement toward more comprehensive fungal sequence databases will continue to increase the robustness of mycobiome characterization in the future. Amplicon sequencing may be advantageous in situations where identification of less abundant fungi is important to the research question, whereas a metagenomics approach could be advantageous toward the study of interkingdom relationships and their impact on health outcomes using a single sequencing experiment.

Data availability statement

Raw sequence files can be found at the NCBI bio project https://www.ncbi.nlm.nih.gov/bioproject/PRJNA903921/. ITS and metagenomic fungal sequence databases can be found at: https://zenodo.org/records/15263682. Metadata and RStudio processing files can be found at: https://github.com/kegle015/Gut-Mycobiome-Maturation-and-Its-Determinants-During-Early-Childhood.

Ethics statement

The studies involving human subjects were approved by the Committee for the Protection of Human Subjects at Children’s Hospital of Philadelphia (CHOP) (IRB 15-012623) and by the Human Research Protection Program at the University of Minnesota (U of MN) (IRB ID STUDY00003037) for the use of de-identified subject clinical and demographic data from CHOP and the use and generation of microbial sequence data from subject fecal samples. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin.

Author contributions

TH: Conceptualization, Validation, Writing – original draft, Writing – review & editing, Data curation, Formal analysis, Investigation, Visualization. SG: Conceptualization, Data curation, Formal analysis, Investigation, Validation, Visualization, Writing – original draft, Writing – review & editing. AD: Formal analysis, Investigation, Writing – review & editing. SH: Data curation, Resources, Writing – review & editing. GA-G: Resources, Writing – review & editing, Formal analysis, Methodology, Software. DG: Methodology, Resources, Writing – review & editing. SM: Resources, Writing – review & editing, Data curation, Project administration. KP: Project administration, Resources, Writing – review & editing. PK: Conceptualization, Formal analysis, Writing – review & editing. MS: Formal analysis, Writing – review & editing. DK: Formal analysis, Writing – review & editing, Conceptualization, Funding acquisition, Project administration. AJ: Conceptualization, Formal analysis, Writing – review & editing. JG: Writing – review & editing, Funding acquisition, Project administration, Resources, Supervision. CG: Funding acquisition, Project administration, Supervision, Writing – review & editing, Conceptualization, Validation, Writing – original draft.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. This research was funded by NIH R21AI139730 (CG, JG, and DK), NIH R01AI121383 (DK and JG), and the University of MN Department of Pediatrics (CG).

Acknowledgments

The authors acknowledge the study participants and their families for their support in this research, as well as study staff for their efforts in recruitment of, and communication with, study participants and their coordination of the clinical aspects of study procedures.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The authors declare that no Gen AI was used in the creation of this manuscript.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb.2025.1539750/full#supplementary-material

References

Ajslev, T. A., Andersen, C. S., Gamborg, M., Sørensen, T. I. A., and Jess, T. (2011). Childhood overweight after establishment of the gut microbiota: the role of delivery mode, pre-pregnancy weight and early administration of antibiotics. Int. J. Obes. 35, 522–529. doi: 10.1038/ijo.2011.27

Crossref Full Text | Google Scholar

Al-Ghalith, G. A., Hillmann, B., Ang, K., Shields-Cutler, R., and Knights, D. (2018). SHI7 is a self-learning pipeline for multipurpose short-read DNA quality. Control 3, e00202–e00217. doi: 10.1128/mSystems.00202-17

PubMed Abstract | Crossref Full Text | Google Scholar

Al-Ghalith, G., and Knights, D. (2017). Knights-lab/BURST: BURST v0.99.4a Technical report. Available online at: https://zenodo.org/record/836859#.YQoxOku_paQ (Accessed May, 2019).

Google Scholar

Amenyogbe, N., Adu-Gyasi, D., Enuameh, Y., Asante, K. P., Konadu, D. G., Kaali, S., et al. (2021). Bacterial and fungal gut community dynamics over the first 5 years of life in predominantly rural communities in Ghana. Front. Microbiol. 12:664407. doi: 10.3389/fmicb.2021.664407

PubMed Abstract | Crossref Full Text | Google Scholar

Antonopoulos, D. A., Huse, S. M., Morrison, H. G., Schmidt, T. M., Sogin, M. L., and Young, V. B. (2009). Reproducible community dynamics of the gastrointestinal microbiota following antibiotic perturbation. Infect. Immun. 77, 2367–2375. doi: 10.1128/IAI.01520-08

PubMed Abstract | Crossref Full Text | Google Scholar

Auchtung, T. A., Stewart, C. J., Smith, D. P., Triplett, E. W., Agardh, D., Hagopian, W. A., et al. (2022). Temporal changes in gastrointestinal fungi and the risk of autoimmunity during early childhood: the TEDDY study. Nat. Commun. 13:3151. doi: 10.1038/s41467-022-30686-w

PubMed Abstract | Crossref Full Text | Google Scholar

Bardanzellu, F., Peroni, D. G., and Fanos, V. (2020). Human breast milk: bioactive components, from stem cells to health outcomes. Curr. Nutr. Rep. 9, 1–13. doi: 10.1007/s13668-020-00303-7

PubMed Abstract | Crossref Full Text | Google Scholar

Boix-Amoros, A., Martinez-Costa, C., Querol, A., Collado, M. C., and Mira, A. (2017). Multiple approaches detect the presence of fungi in human breastmilk samples from healthy mothers. Sci. Rep. 7:13016. doi: 10.1038/s41598-017-13270-x

PubMed Abstract | Crossref Full Text | Google Scholar

Boutin, R. C. T., Sbihi, H., McLaughlin, R. J., Hahn, A. S., Konwar, K. M., Loo, R. S., et al. (2021). Composition and associations of the infant gut fungal microbiota with environmental factors and childhood allergic outcomes. mBio 12:e0339620. doi: 10.1128/mBio.03396-20

PubMed Abstract | Crossref Full Text | Google Scholar

Cho, I., Yamanishi, S., Cox, L., Methé, B. A., Zavadil, J., Li, K., et al. (2012). Antibiotics in early life alter the murine colonic microbiome and adiposity. Nature 488, 621–626. doi: 10.1038/nature11400

PubMed Abstract | Crossref Full Text | Google Scholar

Cox, L. M., Yamanishi, S., Sohn, J., Alekseyenko, A. V., Leung, J. M., Cho, I., et al. (2014). Altering the intestinal microbiota during a critical developmental window has lasting metabolic consequences. Cell 158, 705–721. doi: 10.1016/j.cell.2014.05.052

PubMed Abstract | Crossref Full Text | Google Scholar

Dethlefsen, L., Huse, S., Sogin, M. L., and Relman, D. A. (2008). The pervasive effects of an antibiotic on the human gut microbiota, as revealed by deep 16S rRNA sequencing. PLoS Biol. 6:e280. doi: 10.1371/journal.pbio.0060280

PubMed Abstract | Crossref Full Text | Google Scholar

Dethlefsen, L., and Relman, D. A. (2011). Incomplete recovery and individualized responses of the human distal gut microbiota to repeated antibiotic perturbation. Proc. Natl. Acad. Sci. U.S.A. 108, 4554–4561. doi: 10.1073/pnas.1000087107

PubMed Abstract | Crossref Full Text | Google Scholar

Everard, A., Matamoros, S., Geurts, L., Delzenne, N. M., and Cani, P. D. (2014). Saccharomyces boulardii administration changes gut microbiota and reduces hepatic steatosis, low-grade inflammation, and fat mass in obese and type 2 diabetic db/db mice. mBio 5, e01011–e01014. doi: 10.1128/mBio.01011-14

PubMed Abstract | Crossref Full Text | Google Scholar

Fujimura, K. E., Sitarik, A. R., Havstad, S., Lin, D. L., Levan, S., Fadrosh, D., et al. (2016). Neonatal gut microbiota associates with childhood multisensitized atopy and T cell differentiation. Nat. Med. 22, 1187–1191. doi: 10.1038/nm.4176

PubMed Abstract | Crossref Full Text | Google Scholar

Gohl, D. M., Vangay, P., Garbe, J., MacLean, A., Hauge, A., Becker, A., et al. (2016). Systematic improvement of amplicon marker gene methods for increased accuracy in microbiome studies. Nat. Biotechnol. 34, 942–949. doi: 10.1038/nbt.3601

PubMed Abstract | Crossref Full Text | Google Scholar

Gutierrez, M. W., Mercer, E. M., Moossavi, S., Laforest-Lapointe, I., Reyna, M. E., Becker, A. B., et al. (2023). Maturational patterns of the infant gut mycobiome are associated with early-life body mass index. Cell Rep. Med. 4:100928. doi: 10.1016/j.xcrm.2023.100928

PubMed Abstract | Crossref Full Text | Google Scholar

Hedin, K. A., Rees, V. E., Zhang, H., Kruse, V., Vazquez-Uribe, R., and Sommer, M. O. A. (2022). Effects of broad-spectrum antibiotics on the colonisation of probiotic yeast Saccharomyces boulardii in the murine gastrointestinal tract. Sci. Rep. 12:8862. doi: 10.1038/s41598-022-12806-0

PubMed Abstract | Crossref Full Text | Google Scholar

Heisel, T., Johnson, A. J., Gonia, S., Dillon, A., Skalla, E., Haapala, J., et al. (2022). Bacterial, fungal, and interkingdom microbiome features of exclusively breastfeeding dyads are associated with infant age, antibiotic exposure, and birth mode. Front. Microbiol. 13:1050574. doi: 10.3389/fmicb.2022.1050574

PubMed Abstract | Crossref Full Text | Google Scholar

Heisel, T., Nyaribo, L., Sadowsky, M. J., and Gale, C. A. (2019). Breastmilk and NICU surfaces are potential sources of fungi for infant mycobiomes. Fungal Genet. Biol. 128, 29–35. doi: 10.1016/j.fgb.2019.03.008

PubMed Abstract | Crossref Full Text | Google Scholar

Heisel, T., Podgorski, H., Staley, C. M., Knights, D., Sadowsky, M. J., and Gale, C. A. (2015). Complementary amplicon-based genomic approaches for the study of fungal communities in humans. PLoS One 10:e0116705. doi: 10.1371/journal.pone.0116705

PubMed Abstract | Crossref Full Text | Google Scholar

Iliev, I. D., Funari, V. A., Taylor, K. D., Nguyen, Q., Reyes, C. N., Strom, S. P., et al. (2012). Interactions between commensal fungi and the C-type lectin receptor dectin-1 influence colitis. Science 336, 1314–1317. doi: 10.1126/science.1221789

PubMed Abstract | Crossref Full Text | Google Scholar

Lind, A. L., and Pollard, K. S. (2021). Accurate and sensitive detection of microbial eukaryotes from whole metagenome shotgun sequencing. Microbiome 9:58. doi: 10.1186/s40168-021-01015-y

PubMed Abstract | Crossref Full Text | Google Scholar

Lücking, R., Aime, M. C., Robbertse, B., Miller, A. N., Ariyawansa, H. A., Aoki, T., et al. (2020). Unambiguous identification of fungi: where do we stand and how accurate and precise is fungal DNA barcoding? IMA Fungus 11:14. doi: 10.1186/s43008-020-00033-z

PubMed Abstract | Crossref Full Text | Google Scholar

Mar Rodríguez, M., Pérez, D., Javier Chaves, F., Esteve, E., Marin-Garcia, P., Xifra, G., et al. (2015). Obesity changes the human gut mycobiome. Sci. Rep. 5:14600. doi: 10.1038/srep14600

PubMed Abstract | Crossref Full Text | Google Scholar

Martin, V. M., Virkud, Y. V., Dahan, E., Seay, H. L., Itzkovits, D., Vlamakis, H., et al. (2022). Longitudinal disease-associated gut microbiome differences in infants with food protein-induced allergic proctocolitis. Microbiome 10:154. doi: 10.1186/s40168-022-01322-y

PubMed Abstract | Crossref Full Text | Google Scholar

Mitre, E., Susi, A., Kropp, L. E., Schwartz, D. J., Gorman, G. H., and Nylund, C. M. (2018). Association between use of acid-suppressive medications and antibiotics during infancy and allergic diseases in early childhood. JAMA Pediatr. 172:e180315. doi: 10.1001/jamapediatrics.2018.0315

PubMed Abstract | Crossref Full Text | Google Scholar

Moossavi, S., Fehr, K., Derakhshani, H., Sbihi, H., Robertson, B., Bode, L., et al. (2020). Human milk fungi: environmental determinants and inter-kingdom associations with milk bacteria in the CHILD cohort study. BMC Microbiol. 20:146. doi: 10.1186/s12866-020-01829-0

PubMed Abstract | Crossref Full Text | Google Scholar

Murphy, R., Stewart, A. W., Braithwaite, I., Beasley, R., Hancox, R. J., Mitchell, E. A., et al. (2014). Antibiotic treatment during infancy and increased body mass index in boys: an international cross-sectional study. Int. J. Obes. 38, 1115–1119. doi: 10.1038/ijo.2013.218

PubMed Abstract | Crossref Full Text | Google Scholar

Nash, A. K., Auchtung, T. A., Wong, M. C., Smith, D. P., Gesell, J. R., Ross, M. C., et al. (2017). The gut mycobiome of the human microbiome project healthy cohort. Microbiome 5:153. doi: 10.1186/s40168-017-0373-4

PubMed Abstract | Crossref Full Text | Google Scholar

Nilsson, R. H., Anslan, S., Bahram, M., Wurzbacher, C., Baldrian, P., and Tedersoo, L. (2019). Mycobiome diversity: high-throughput sequencing and identification of fungi. Nat. Rev. Microbiol. 17, 95–109. doi: 10.1038/s41579-018-0116-y

PubMed Abstract | Crossref Full Text | Google Scholar

Olm, M. R., Dahan, D., Carter, M. M., Merrill, B. D., Yu, F. B., Jain, S., et al. (2022). Robust variation in infant gut microbiome assembly across a spectrum of lifestyles. Science 376, 1220–1223. doi: 10.1126/science.abj2972

PubMed Abstract | Crossref Full Text | Google Scholar

Ott, S. J., Kühbacher, T., Musfeldt, M., Rosenstiel, P., Hellmig, S., Rehman, A., et al. (2008). Fungi and inflammatory bowel diseases: alterations of composition and diversity. Scand. J. Gastroenterol. 43, 831–841. doi: 10.1080/00365520801935434

PubMed Abstract | Crossref Full Text | Google Scholar

Palmer, J. M., Jusino, M. A., Banik, M. T., and Lindner, D. L. (2018). Non-biological synthetic spike-in controls and the AMPtk software pipeline improve mycobiome data. PeerJ 6:e4925. doi: 10.7717/peerj.4925

PubMed Abstract | Crossref Full Text | Google Scholar

Parfrey, L. W., Walters, W. A., Lauber, C. L., Clemente, J. C., Berg-Lyons, D., Teiling, C., et al. (2014). Communities of microbial eukaryotes in the mammalian gut within the context of environmental eukaryotic diversity. Front. Microbiol. 5:298. doi: 10.3389/fmicb.2014.00298

PubMed Abstract | Crossref Full Text | Google Scholar

Robinson, C. J., and Young, V. B. (2010). Antibiotic administration alters the community structure of the gastrointestinal micobiota. Gut Microbes 1, 279–284. doi: 10.4161/gmic.1.4.12614

PubMed Abstract | Crossref Full Text | Google Scholar

Schei, K., Avershina, E., Øien, T., Rudi, K., Follestad, T., Salamati, S., et al. (2017). Early gut mycobiota and mother-offspring transfer. Microbiome 5:107. doi: 10.1186/s40168-017-0319-x

PubMed Abstract | Crossref Full Text | Google Scholar

Schoch, C. L., Seifert, K. A., Huhndorf, S., Robert, V., Spouge, J. L., Levesque, C. A., et al. (2012). Nuclear ribosomal internal transcribed spacer (ITS) region as a universal DNA barcode marker for Fungi. Proc. Natl. Acad. Sci. U.S.A. 109, 6241–6246. doi: 10.1073/pnas.1117018109

PubMed Abstract | Crossref Full Text | Google Scholar

Sokol, H., Leducq, V., Aschard, H., Pham, H.-P., Jegou, S., Landman, C., et al. (2017). Fungal microbiota dysbiosis in IBD. Gut 66, 1039–1048. doi: 10.1136/gutjnl-2015-310746

Crossref Full Text | Google Scholar

Stewart, C. J., Ajami, N. J., O’Brien, J. L., Hutchinson, D. S., Smith, D. P., Wong, M. C., et al. (2018). Temporal development of the gut microbiome in early childhood from the TEDDY study. Nature 562, 583–588. doi: 10.1038/s41586-018-0617-x

PubMed Abstract | Crossref Full Text | Google Scholar

Stražar, M., Temba, G. S., Vlamakis, H., Kullaya, V. I., Lyamuya, F., Mmbaga, B. T., et al. (2021). Gut microbiome-mediated metabolism effects on immunity in rural and urban African populations. Nat. Commun. 12:4845. doi: 10.1038/s41467-021-25213-2

PubMed Abstract | Crossref Full Text | Google Scholar

van Tilburg Bernardes, E., Pettersen, V. K., Gutierrez, M. W., Laforest-Lapointe, I., Jendzjowsky, N. G., Cavin, J.-B., et al. (2020). Intestinal fungi are causally implicated in microbiome assembly and immune development in mice. Nat. Commun. 11:2577. doi: 10.1038/s41467-020-16431-1

PubMed Abstract | Crossref Full Text | Google Scholar

Vatanen, T., Franzosa, E. A., Schwager, R., Tripathi, S., Arthur, T. D., Vehik, K., et al. (2018). The human gut microbiome in early-onset type 1 diabetes from the TEDDY study. Nature 562, 589–594. doi: 10.1038/s41586-018-0620-2

PubMed Abstract | Crossref Full Text | Google Scholar

Ventin-Holmberg, R., Saqib, S., Korpela, K., Nikkonen, A., Peltola, V., Salonen, A., et al. (2022). The effect of antibiotics on the infant gut fungal microbiota. J. Fungi 8:328. doi: 10.3390/jof8040328

PubMed Abstract | Crossref Full Text | Google Scholar

Wampach, L., Heintz-Buschart, A., Hogan, A., Muller, E. E. L., Narayanasamy, S., Laczny, C. C., et al. (2017). Colonization and succession within the human gut microbiome by archaea, bacteria, and microeukaryotes during the first year of life. Front. Microbiol. 8:738. doi: 10.3389/fmicb.2017.00738

PubMed Abstract | Crossref Full Text | Google Scholar