American Gut: an Open Platform for Citizen Science Microbiome Research

Abstract

Although much work has linked the human microbiome to specific phenotypes and lifestyle variables, data from different projects have been challenging to integrate and the extent of microbial and molecular diversity in human stool remains unknown. Using standardized protocols from the Earth Microbiome Project and sample contributions from over 10,000 citizen-scientists, together with an open research network, we compare human microbiome specimens primarily from the United States, United Kingdom, and Australia to one another and to environmental samples. Our results show an unexpected range of beta-diversity in human stool microbiomes compared to environmental samples; demonstrate the utility of procedures for removing the effects of overgrowth during room-temperature shipping for revealing phenotype correlations; uncover new molecules and kinds of molecular communities in the human stool metabolome; and examine emergent associations among the microbiome, metabolome, and the diversity of plants that are consumed (rather than relying on reductive categorical variables such as veganism, which have little or no explanatory power). We also demonstrate the utility of the living data resource and cross-cohort comparison to confirm existing associations between the microbiome and psychiatric illness and to reveal the extent of microbiome change within one individual during surgery, providing a paradigm for open microbiome research and education. IMPORTANCE We show that a citizen science, self-selected cohort shipping samples through the mail at room temperature recaptures many known microbiome results from clinically collected cohorts and reveals new ones. Of particular interest is integrating n = 1 study data with the population data, showing that the extent of microbiome change after events such as surgery can exceed differences between distinct environmental biomes, and the effect of diverse plants in the diet, which we confirm with untargeted metabolomics on hundreds of samples.

FIG 1

Population characteristics. (A) Participants across the world have sent in samples to American Gut, although the primary geographic regions of participation are in North America and the United Kingdom; the report that a participant receives is depicted. (B) The primary sample breakdown for subsequent analyses. Red denotes the reasons that samples were removed. (C) Between the two largest populations, the United States (n = 6,634) and the United Kingdom (n = 2,071), we observe a significant difference in alpha-diversity. (D) In a meta-analysis, the largely industrialized population that makes up American Gut exhibits significant differential abundances compared to nonindustrialized populations.

FIG 2

Blooms and effect sizes. (A) The fraction of 16S reads that recruit to bloom reads defined by Amir et al. (15) is strongly associated with the likelihood for microbial growth under aerobic culture conditions on rich medium. (B) Overlap of mass spectral features (consensus MS/MS cluster nodes; see Materials and Methods, “Molecular networking”) between AGP samples and blooms. (C) Unweighted UniFrac effect sizes. The inset shows the correlation of effect sizes when including or excluding the bloom 16S reads (Pearson r = 0.91, P = 3.76 × 10⁻⁵⁷). (D) Weighted UniFrac effect sizes. The inset shows the correlation of the effect sizes when including or excluding bloom 16S reads (Pearson r = 0.42, P = 1.71 × 10⁻⁶); the outlier is the 16S bloom fraction of the sample.

FIG 3

OTU and beta-diversity novelty. (A) The AGP data placed into the context of extant microbial diversity at a global scale. (B) A phylogenetic tree showing the diversity spanned by the AGP and the HMP in the context of Greengenes and the EMP. (C and D) sOTU novelty over increasing numbers of samples in the AGP (C); the AGP appears to have begun to reach saturation and is contrasted with the data from the work of Yatsunenko et al. (6) (D), which, unlike the AGP, had extremely deep sequencing per sample. (E) The minimum observed UniFrac distance between samples over increasing numbers of samples for the AGP and the HMP; the inset is from 0 to 500 samples. (F) An AGP “trading card” of an sOTU of interest (shown in full in Fig. S2).

FIG 4

Temporal and spatial patterns. (A) Five hundred sixty-five individuals had multiple samples. Distances between samples within an individual shown at 1 month, 2 months, etc., out to over 1 year; between-subject distances are shown as BSD. Even at 1 year, the median distance between a participant’s samples is less than the median between-participant distance. (B) Within the United States, spatial processes of sOTUs appear driven by stochastic processes, as few sOTUs exhibit spatial autocorrelation (Moran’s I) on the full data set or partitions (e.g., participants older than 20 years). (C) Distance-decay relationship for Bray-Curtis dissimilarities between subject pairs that are within a 100-km (great-circle distance) neighborhood radius of one another (Mantel test r = 0.036, adjusted P = 0.03). To avoid the overplotting associated with visualization of the more than 3.4 × 10⁵ pairwise comparisons, we visualized this relationship using two-dimensional frequency bins; darker colors indicate higher-frequency bins. Solid lines represent fits from linear models to raw data. The inset shows the largest radius (i.e., the contiguous United States). Axes are the same as in the large panel. (D) Mantel correlogram of estimated Mantel r correlations, significance of distance-decay relationships, and neighborhood size (x axis). Filled points represent neighborhood sizes for which distance-decay relationships were significant (adjusted P values < 0.05). (E) Characterizing a large bowel resection using the AGP, the EMP, a hunter-gatherer population, and ICU patients in an unweighted UniFrac principal-coordinate plot. A state change was observed in the resulting microbial community. The change in the microbial community immediately following surgery is the same as the distance between a marine sediment sample and a plant rhizosphere sample.

FIG 5

Diversity of plants in a diet. (A) Procrustes analysis of fecal samples from n = 1,596 individuals using principal components of the VioScreen FFQ responses and principal coordinates of the unweighted UniFrac distances (M² = 0.988) colored by diet; Procrustes tests the fit of one ordination space to another. PCA shows grouping by diets such as vegan, suggesting that self-reported diet type is consistent with differences in micronutrients and macronutrients as recorded by the FFQ; however, these dietary differences do not explain relationships between the samples in 16S space. (B) The full AGP data set, including skin and oral samples, through unweighted UniFrac and principal-coordinate analysis, highlighting a lack of apparent clustering by diet type. (C and D) Dietary conjugated linoleic acid levels as reported by the FFQ between the extremes of plant diversity consumption (C) and the levels of CLA observed by HPLC-MS (D). (E) Differential abundances of sOTUs (showing the most specific taxon name per sOTU) between those who eat fewer than 10 plants per week and those who eat over 30 per week. (F) The molecules linoleic acid (LA) and conjugated linoleic acid (CLA) (only trans-, trans-isomers are shown) were found to comprise the octadecadienoic acid found to be the key feature in this difference in number of plants consumed.

FIG 6

(A) Compound occurrence frequency plot. Examples of compounds originating from food (piperine, black pepper alkaloid), host (stercobilin, heme catabolism product), and bacterial activity (lithocholic acid, microbially modified bile acid) or exogenous compounds such as antibiotics (rifaximin) or other drugs (lisinopril, high blood pressure medication) are shown. (B to E) Alpha- and beta-diversity assessments of antibiotic (B and C) and plant (D and E) cohorts; insets depict minimum observed beta-diversity over increasing samples.