A distinct Fusobacterium nucleatum clade dominates the colorectal cancer niche.

Fusobacterium nucleatum (Fn), a bacterium present in the human oral cavity and rarely found in the lower gastrointestinal tract of healthy individuals1, is enriched in human colorectal cancer (CRC) tumours2-5. High intratumoural Fn loads are associated with recurrence, metastases and poorer patient prognosis5-8. Here, to delineate Fn genetic factors facilitating tumour colonization, we generated closed genomes for 135 Fn strains; 80 oral strains from individuals without cancer and 55 unique cancer strains cultured from tumours from 51 patients with CRC. Pangenomic analyses identified 483 CRC-enriched genetic factors. Tumour-isolated strains predominantly belong to Fn subspecies animalis (Fna). However, genomic analyses reveal that Fna, considered a single subspecies, is instead composed of two distinct clades (Fna C1 and Fna C2). Of these, only Fna C2 dominates the CRC tumour niche. Inter-Fna analyses identified 195 Fna C2-associated genetic factors consistent with increased metabolic potential and colonization of the gastrointestinal tract. In support of this, Fna C2-treated mice had an increased number of intestinal adenomas and altered metabolites. Microbiome analysis of human tumour tissue from 116 patients with CRC demonstrated Fna C2 enrichment. Comparison of 62 paired specimens showed that only Fna C2 is tumour enriched compared to normal adjacent tissue. This was further supported by metagenomic analysis of stool samples from 627 patients with CRC and 619 healthy individuals. Collectively, our results identify the Fna clade bifurcation, show that specifically Fna C2 drives the reported Fn enrichment in human CRC and reveal the genetic underpinnings of pathoadaptation of Fna C2 to the CRC niche.

Effect of the intratumoral microbiota on spatial and cellular heterogeneity in cancer.

The tumour-associated microbiota is an intrinsic component of the tumour microenvironment across human cancer types1,2. Intratumoral host-microbiota studies have so far largely relied on bulk tissue analysis1-3, which obscures the spatial distribution and localized effect of the microbiota within tumours. Here, by applying in situ spatial-profiling technologies4 and single-cell RNA sequencing5 to oral squamous cell carcinoma and colorectal cancer, we reveal spatial, cellular and molecular host-microbe interactions. We adapted 10x Visium spatial transcriptomics to determine the identity and in situ location of intratumoral microbial communities within patient tissues. Using GeoMx digital spatial profiling6, we show that bacterial communities populate microniches that are less vascularized, highly immuno­suppressive and associated with malignant cells with lower levels of Ki-67 as compared to bacteria-negative tumour regions. We developed a single-cell RNA-sequencing method that we name INVADEseq (invasion-adhesion-directed expression sequencing) and, by applying this to patient tumours, identify cell-associated bacteria and the host cells with which they interact, as well as uncovering alterations in transcriptional pathways that are involved in inflammation, metastasis, cell dormancy and DNA repair. Through functional studies, we show that cancer cells that are infected with bacteria invade their surrounding environment as single cells and recruit myeloid cells to bacterial regions. Collectively, our data reveal that the distribution of the microbiota within a tumour is not random; instead, it is highly organized in microniches with immune and epithelial cell functions that promote cancer progression.

The Regulation of Nucleic Acid Vaccine Responses by the Microbiome.

Nucleic acid vaccines, including both RNA and DNA platforms, are key technologies that have considerable promise in combating both infectious disease and cancer. However, little is known about the extrinsic factors that regulate nucleic acid vaccine responses and which may determine their effectiveness. The microbiome is recognized as a significant regulator of immune development and response, whose role in regulating some traditional vaccine platforms has recently been discovered. Using germ-free and specific pathogen-free mouse models in combination with different protein, DNA, and mRNA vaccine regimens, we demonstrate that the microbiome is a significant regulator of nucleic acid vaccine immunogenicity. Although the presence of the microbiome enhances CD8+ T cell responses to mRNA lipid nanoparticle immunization, the microbiome suppresses Ig and CD4+ T cell responses to DNA-prime, DNA-protein-boost immunization, indicating contrasting roles for the microbiome in the regulation of these different nucleic acid vaccine platforms. In the case of mRNA lipid nanoparticle vaccination, germ-free mice display reduced dendritic cell/macrophage activation that may underlie the deficient vaccine response. Our study identifies the microbiome as a relevant determinant of nucleic acid vaccine response with implications for continued therapeutic development and deployment of these vaccines.

phippery: a software suite for PhIP-Seq data analysis.

SUMMARY: We present the phippery software suite for analyzing data from phage display methods that use immunoprecipitation and deep sequencing to capture antibody binding to peptides, often referred to as PhIP-Seq. It has three main components that can be used separately or in conjunction: (i) a Nextflow pipeline, phip-flow, to process raw sequencing data into a compact, multidimensional dataset format and allows for end-to-end automation of reproducible workflows. (ii) a Python API, phippery, which provides interfaces for tasks such as count normalization, enrichment calculation, multidimensional scaling, and more, and (iii) a Streamlit application, phip-viz, as an interactive interface for visualizing the data as a heatmap in a flexible manner. AVAILABILITY AND IMPLEMENTATION: All software packages are publicly available under the MIT License. The phip-flow pipeline: https://github.com/matsengrp/phip-flow. The phippery library: https://github.com/matsengrp/phippery. The phip-viz Streamlit application: https://github.com/matsengrp/phip-viz.

In silico benchmarking of metagenomic tools for coding sequence detection reveals the limits of sensitivity and precision.

BACKGROUND: High-throughput sequencing can establish the functional capacity of a microbial community by cataloging the protein-coding sequences (CDS) present in the metagenome of the community. The relative performance of different computational methods for identifying CDS from whole-genome shotgun sequencing is not fully established. RESULTS: Here we present an automated benchmarking workflow, using synthetic shotgun sequencing reads for which we know the true CDS content of the underlying communities, to determine the relative performance (sensitivity, positive predictive value or PPV, and computational efficiency) of different metagenome analysis tools for extracting the CDS content of a microbial community. Assembly-based methods are limited by coverage depth, with poor sensitivity for CDS at < 5X depth of sequencing, but have excellent PPV. Mapping-based techniques are more sensitive at low coverage depths, but can struggle with PPV. We additionally describe an expectation maximization based iterative algorithmic approach which we show to successfully improve the PPV of a mapping based technique while retaining improved sensitivity and computational efficiency. CONCLUSION: Our benchmarking approach reveals the trade-offs of assembly versus alignment-based approaches and the relative performance of specific implementations when one wishes to extract the protein coding capacity of microbial communities.

Metagenomic Analysis Reveals Clinical SARS-CoV-2 Infection and Bacterial or Viral Superinfection and Colonization.

BACKGROUND: More than 2 months separated the initial description of SARS-CoV-2 and discovery of its widespread dissemination in the United States. Despite this lengthy interval, implementation of specific quantitative reverse transcription (qRT)-PCR-based SARS-CoV-2 tests in the US has been slow, and testing is still not widely available. Metagenomic sequencing offers the promise of unbiased detection of emerging pathogens, without requiring prior knowledge of the identity of the responsible agent or its genomic sequence. METHODS: To evaluate metagenomic approaches in the context of the current SARS-CoV-2 epidemic, laboratory-confirmed positive and negative samples from Seattle, WA were evaluated by metagenomic sequencing, with comparison to a 2019 reference genomic database created before the emergence of SARS-CoV-2. RESULTS: Within 36 h our results showed clear identification of a novel human Betacoronavirus, closely related to known Betacoronaviruses of bats, in laboratory-proven cases of SARS-CoV-2. A subset of samples also showed superinfection or colonization with human parainfluenza virus 3 or Moraxella species, highlighting the need to test directly for SARS-CoV-2 as opposed to ruling out an infection using a viral respiratory panel. Samples negative for SARS-CoV-2 by RT-PCR were also negative by metagenomic analysis, and positive for Rhinovirus A and C. Unlike targeted SARS-CoV-2 qRT-PCR testing, metagenomic analysis of these SARS-CoV-2 negative samples identified candidate etiological agents for the patients' respiratory symptoms. CONCLUSION: Taken together, these results demonstrate the value of metagenomic analysis in the monitoring and response to this and future viral pandemics.

Global divergence of the human follicle mite Demodex folliculorum: Persistent associations between host ancestry and mite lineages.

Microscopic mites of the genus Demodex live within the hair follicles of mammals and are ubiquitous symbionts of humans, but little molecular work has been done to understand their genetic diversity or transmission. Here we sampled mite DNA from 70 human hosts of diverse geographic ancestries and analyzed 241 sequences from the mitochondrial genome of the species Demodex folliculorum. Phylogenetic analyses recovered multiple deep lineages including a globally distributed lineage common among hosts of European ancestry and three lineages that primarily include hosts of Asian, African, and Latin American ancestry. To a great extent, the ancestral geography of hosts predicted the lineages of mites found on them; 27% of the total molecular variance segregated according to the regional ancestries of hosts. We found that D. folliculorum populations are stable on an individual over the course of years and that some Asian and African American hosts maintain specific mite lineages over the course of years or generations outside their geographic region of birth or ancestry. D. folliculorum haplotypes were much more likely to be shared within families and between spouses than between unrelated individuals, indicating that transmission requires close contact. Dating analyses indicated that D. folliculorum origins may predate modern humans. Overall, D. folliculorum evolution reflects ancient human population divergences, is consistent with an out-of-Africa dispersal hypothesis, and presents an excellent model system for further understanding the history of human movement.

Rapid evolution of the human gut virome.

Humans are colonized by immense populations of viruses, which metagenomic analysis shows are mostly unique to each individual. To investigate the origin and evolution of the human gut virome, we analyzed the viral community of one adult individual over 2.5 y by extremely deep metagenomic sequencing (56 billion bases of purified viral sequence from 24 longitudinal fecal samples). After assembly, 478 well-determined contigs could be identified, which are inferred to correspond mostly to previously unstudied bacteriophage genomes. Fully 80% of these types persisted throughout the duration of the 2.5-y study, indicating long-term global stability. Mechanisms of base substitution, rates of accumulation, and the amount of variation varied among viral types. Temperate phages showed relatively lower mutation rates, consistent with replication by accurate bacterial DNA polymerases in the integrated prophage state. In contrast, Microviridae, which are lytic bacteriophages with single-stranded circular DNA genomes, showed high substitution rates (>10(-5) per nucleotide each day), so that sequence divergence over the 2.5-y period studied approached values sufficient to distinguish new viral species. Longitudinal changes also were associated with diversity-generating retroelements and virus-encoded Clustered Regularly Interspaced Short Palindromic Repeats arrays. We infer that the extreme interpersonal diversity of human gut viruses derives from two sources, persistence of a small portion of the global virome within the gut of each individual and rapid evolution of some long-term virome members.

Hypervariable loci in the human gut virome.

Genetic variation is critical in microbial immune evasion and drug resistance, but variation has rarely been studied in complex heterogeneous communities such as the human microbiome. To begin to study natural variation, we analyzed DNA viruses present in the lower gastrointestinal tract of 12 human volunteers by determining 48 billion bases of viral DNA sequence. Viral genomes mostly showed low variation, but 51 loci of ∼100 bp showed extremely high variation, so that up to 96% of the viral genomes encoded unique amino acid sequences. Some hotspots of hypervariation were in genes homologous to the bacteriophage BPP-1 viral tail-fiber gene, which is known to be hypermutagenized by a unique reverse-transcriptase (RT)-based mechanism. Unexpectedly, other hypervariable loci in our data were in previously undescribed gene types, including genes encoding predicted Ig-superfamily proteins. Most of the hypervariable loci were linked to genes encoding RTs of a single clade, which we find is the most abundant clade among gut viruses but only a minor component of bacterial RT populations. Hypervariation was targeted to 5'-AAY-3' asparagine codons, which allows maximal chemical diversification of the encoded amino acids while avoiding formation of stop codons. These findings document widespread targeted hypervariation in the human gut virome, identify previously undescribed types of genes targeted for hypervariation, clarify association with RT gene clades, and motivate studies of hypervariation in the full human microbiome.

Admixture and recombination among Toxoplasma gondii lineages explain global genome diversity.

Toxoplasma gondii is a highly successful protozoan parasite that infects all warm-blooded animals and causes severe disease in immunocompromised and immune-naïve humans. It has an unusual global population structure: In North America and Europe, isolated strains fall predominantly into four largely clonal lineages, but in South America there is great genetic diversity and the North American clonal lineages are rarely found. Genetic variation between Toxoplasma strains determines differences in virulence, modulation of host-signaling pathways, growth, dissemination, and disease severity in mice and likely in humans. Most studies on Toxoplasma genetic variation have focused on either a few loci in many strains or low-resolution genome analysis of three clonal lineages. We use whole-genome sequencing to identify a large number of SNPs between 10 Toxoplasma strains from Europe and North and South America. These were used to identify haplotype blocks (genomic regions) shared between strains and construct a Toxoplasma haplotype map. Additional SNP analysis of RNA-sequencing data of 26 Toxoplasma strains, representing global diversity, allowed us to construct a comprehensive genealogy for Toxoplasma gondii that incorporates sexual recombination. These data show that most current isolates are recent recombinants and cannot be easily grouped into a limited number of haplogroups. A complex picture emerges in which some genomic regions have not been recently exchanged between any strains, and others recently spread from one strain to many others.

Complete mitochondrial genomes of the human follicle mites Demodex brevis and D. folliculorum: novel gene arrangement, truncated tRNA genes, and ancient divergence between species.

BACKGROUND: Follicle mites of the genus Demodex are found on a wide diversity of mammals, including humans; surprisingly little is known, however, about the evolution of this association. Additional sequence information promises to facilitate studies of Demodex variation within and between host species. Here we report the complete mitochondrial genome sequences of two species of Demodex known to live on humans--Demodex brevis and D. folliculorum--which are the first such genomes available for any member of the genus. We analyzed these sequences to gain insight into the evolution of mitochondrial genomes within the Acariformes. We also used relaxed molecular clock analyses, based on alignments of mitochondrial proteins, to estimate the time of divergence between these two species. RESULTS: Both Demodex genomes shared a novel gene order that differs substantially from the ancestral chelicerate pattern, with transfer RNA (tRNA) genes apparently having moved much more often than other genes. Mitochondrial tRNA genes of both species were unusually short, with most of them unable to encode tRNAs that could fold into the canonical cloverleaf structure; indeed, several examples lacked both D- and T-arms. Finally, the high level of sequence divergence observed between these species suggests that these two lineages last shared a common ancestor no more recently than about 87 mya. CONCLUSIONS: Among Acariformes, rearrangements involving tRNA genes tend to occur much more often than those involving other genes. The truncated tRNA genes observed in both Demodex species would seem to require the evolution of extensive tRNA editing capabilities and/or coevolved interacting factors. The molecular machinery necessary for these unusual tRNAs to function might provide an avenue for developing treatments of skin disorders caused by Demodex. The deep divergence time estimated between these two species sets a lower bound on the time that Demodex have been coevolving with their mammalian hosts, and supports the hypothesis that there was an early split within the genus Demodex into species that dwell in different skin microhabitats.

The human gut virome: inter-individual variation and dynamic response to diet.

Immense populations of viruses are present in the human gut and other body sites. Understanding the role of these populations (the human "virome") in health and disease requires a much deeper understanding of their composition and dynamics in the face of environmental perturbation. Here, we investigate viromes from human subjects on a controlled feeding regimen. Longitudinal fecal samples were analyzed by metagenomic sequencing of DNA from virus-like particles (VLP) and total microbial communities. Assembly of 336 Mb of VLP sequence yielded 7175 contigs, many identifiable as complete or partial bacteriophage genomes. Contigs were rich in viral functions required in lytic and lysogenic growth, as well as unexpected functions such as viral CRISPR arrays and genes for antibiotic resistance. The largest source of variance among virome samples was interpersonal variation. Parallel deep-sequencing analysis of bacterial populations showed covaration of the virome with the larger microbiome. The dietary intervention was associated with a change in the virome community to a new state, in which individuals on the same diet converged. Thus these data provide an overview of the composition of the human gut virome and associate virome structure with diet.

Multi-cohort shotgun metagenomic analysis of oral and gut microbiota overlap in healthy adults.

The multitude of barriers between the mouth and colon may eliminate swallowed oral bacteria. Ascertaining the presence of the same bacteria in the mouth and colon is methodologically challenging partly because 16S rRNA gene sequencing - the most commonly used method to characterize the human microbiota - has low confidence in taxonomic assignments deeper than genus for most bacteria. As different species of the same genus can have low-level variation across the same 16S rRNA gene region, shotgun sequencing is needed to identify a true overlap. We analyzed a curated, multi-cohort, shotgun metagenomic database with species-level taxonomy and clade-specific marker genes to fill this knowledge gap. Using 500 paired fecal/oral (4 oral sites) samples from 4 healthy adult cohorts, we found a minute overlap between the two niches. Comparing marker genes between paired oral and fecal samples with species-level overlap, the pattern of overlap in only 7 individuals was consistent with same-strain colonization. These findings argue against ectopic colonization of oral bacteria in the distal gut in healthy adults.

Scalable single-cell profiling of chromatin modifications with sciCUT&Tag.

Cleavage under targets and tagmentation (CUT&Tag) is an antibody-directed in situ chromatin profiling strategy that is rapidly replacing immune precipitation-based methods, such as chromatin immunoprecipitation-sequencing. The efficiency of the method enables chromatin profiling in single cells but is limited by the numbers of cells that can be profiled. Here, we describe a combinatorial barcoding strategy for CUT&Tag that harnesses a nanowell dispenser for simple, high-resolution, high-throughput, single-cell chromatin profiling. In this single-cell combinatorial indexing CUT&Tag (sciCUT&Tag) protocol, lightly cross-linked nuclei are bound to magnetic beads and incubated with primary and secondary antibodies in bulk and then arrayed in a 96-well plate for a first round of cellular indexing by antibody-directed Tn5 tagmentation. The sample is then repooled, mixed and arrayed across 5,184 nanowells at a density of 12-24 nuclei per well for a second round of cellular indexing during PCR amplification of the sequencing-ready library. This protocol can be completed in 1.5 days by a research technician, and we illustrate the optimized protocol by profiling histone modifications associated with developmental gene repression (H3K27me3) as well as transcriptional activation (H3K4me1-2-3) in human peripheral blood mononuclear cells and use single-nucleotide polymorphisms to facilitate collision removal. We have also used sciCUT&Tag for simultaneous profiling of multiple chromatin epitopes in single cells. The reduced cost, improved resolution and scalability of sciCUT&Tag make it an attractive platform to profile chromatin features in single cells.

Species- and subspecies-level characterization of health-associated bacterial consortia that colonize the human gut during infancy.

BACKGROUND: The human gut microbiome develops rapidly during infancy, a key window of development coinciding with the maturation of the adaptive immune system. However, little is known about the microbiome growth dynamics over the first few months of life and whether there are any generalizable patterns across human populations. We performed metagenomic sequencing on stool samples (n = 94) from a cohort of infants (n = 15) at monthly intervals in the first 6 months of life, augmenting our dataset with seven published studies for a total of 4,441 metagenomes from 1,162 infants. RESULTS: Strain-level de novo analysis was used to identify 592 of the most abundant organisms in the infant gut microbiome. Previously unrecognized consortia were identified which exhibited highly correlated abundances across samples and were composed of diverse species spanning multiple genera. Analysis of a published cohort of infants with cystic fibrosis identified one such novel consortium of diverse Enterobacterales which was positively correlated with weight gain. While all studies showed an increased community stability during the first year of life, microbial dynamics varied widely in the first few months of life, both by study and by individual. CONCLUSION: By augmenting published metagenomic datasets with data from a newly established cohort, we were able to identify novel groups of organisms that are correlated with measures of robust human development. We hypothesize that the presence of these groups may impact human health in aggregate in ways that individual species may not in isolation.

Colorectal cancer-associated bacteria are broadly distributed in global microbiomes and drivers of precancerous change.

The gut microbiome is implicated in the pathogenesis of colorectal cancer (CRC), but the full scope of this dialogue is unknown. Here we aimed to define the scale and membership of the body of CRC- and health-associated gut bacteria in global populations. We performed a microbiome-CRC correlation analysis of published ultra-deep shotgun metagenomic sequencing data from global microbiome surveys, utilizing a de novo (reference-agnostic) gene-level clustering approach to identify protein-coding co-abundant gene (CAGs) clusters. We link an unprecedented ~ 23-40% of gut bacteria to CRC or health, split nearly evenly as CRC- or health-associated. These microbes encode 2319 CAGs encompassing 427,261 bacterial genes significantly enriched or depleted in CRC. We identified many microbes that had not previously been linked to CRC, thus expanding the scope of "known unknowns" of CRC-associated microbes. We performed an agnostic CAG-based screen of bacterial isolates and validated predicted effects of previously unimplicated bacteria in preclinical models, in which we observed differential induction of precancerous adenomas and field effects. Single-cell RNA sequencing disclosed microbiome-induced senescence-associated gene expression signatures in discrete colonic populations including fibroblasts. In organoid co-cultures, primary colon fibroblasts from mice with microbiomes promoted significantly greater growth than fibroblasts from microbiome-depleted mice. These results offer proof-of-principle for gene-level metagenomic analysis enabling discovery of microbiome links to health and demonstrate that the microbiome can drive precancer states, thereby potentially revealing novel cancer prevention opportunities.

VMAP: Vaginal Microbiome Atlas during Pregnancy.

Objectives: To enable interactive visualization of the vaginal microbiome across the pregnancy and facilitate discovery of novel insights and generation of new hypotheses. Material and Methods: Vaginal Microbiome Atlas during Pregnancy (VMAP) was created with R shiny to generate visualizations of structured vaginal microbiome data from multiple studies. Results: VMAP (http://vmapapp.org) visualizes 3880 vaginal microbiome samples of 1402 pregnant individuals from 11 studies, aggregated via open-source tool MaLiAmPi. Visualized features include diversity measures, VALENCIA community state types, and composition (phylotypes, taxonomy) that can be filtered by various categories. Discussion: This work represents one of the largest and most geographically diverse aggregations of the vaginal microbiome in pregnancy to date and serves as a user-friendly resource to further analyze vaginal microbiome data and better understand pregnancies and associated outcomes. Conclusion: VMAP can be obtained from https://github.com/msirota/vmap.git and is currently deployed as an online app for non-R users.

Microbiome preterm birth DREAM challenge: Crowdsourcing machine learning approaches to advance preterm birth research.

Every year, 11% of infants are born preterm with significant health consequences, with the vaginal microbiome a risk factor for preterm birth. We crowdsource models to predict (1) preterm birth (PTB; <37 weeks) or (2) early preterm birth (ePTB; <32 weeks) from 9 vaginal microbiome studies representing 3,578 samples from 1,268 pregnant individuals, aggregated from public raw data via phylogenetic harmonization. The predictive models are validated on two independent unpublished datasets representing 331 samples from 148 pregnant individuals. The top-performing models (among 148 and 121 submissions from 318 teams) achieve area under the receiver operator characteristic (AUROC) curve scores of 0.69 and 0.87 predicting PTB and ePTB, respectively. Alpha diversity, VALENCIA community state types, and composition are important features in the top-performing models, most of which are tree-based methods. This work is a model for translation of microbiome data into clinically relevant predictive models and to better understand preterm birth.

INVADEseq to identify cell-adherent or invasive bacteria and the associated host transcriptome at single-cell-level resolution.

Single-cell RNA sequencing (scRNAseq) technologies have been beneficial in revealing and describing cellular heterogeneity within mammalian tissues, including solid tumors. However, many of these techniques apply poly(A) selection of RNA, and thus have primarily focused on determining the gene signatures of eukaryotic cellular components of the tumor microenvironment. Microbiome analysis has revealed the presence of microbial ecosystems, including bacteria and fungi, within human tumor tissues from major cancer types. Imaging data have revealed that intratumoral bacteria may be located within epithelial and immune cell types. However, as bacterial RNA typically lacks a poly(A) tail, standard scRNAseq approaches have limited ability to capture this microbial component of the tumor microenvironment. To overcome this, we describe the invasion-adhesion-directed expression sequencing (INVADEseq) approach, whereby we adapt 10x Genomics 5' scRNAseq protocol by introducing a primer that targets a conserved region of the bacterial 16S ribosomal RNA gene in addition to the standard primer for eukaryotic poly(A) RNA selection. This 'add-on' approach enables the generation of eukaryotic and bacterial DNA libraries at eukaryotic single-cell level resolution, utilizing the 10x barcode to identify single cells with intracellular bacteria. The INVADEseq method takes 30 h to complete, including tissue processing, sequencing and computational analysis. As an output, INVADEseq has shown to be a reliable tool in human cancer cell lines and patient tumor specimens by detecting the proportion of human cells that harbor bacteria and the identities of human cells and intracellular bacteria, along with identifying host transcriptional programs that are modulated on the basis of associated bacteria.

MaLiAmPi enables generalizable and taxonomy-independent microbiome features from technically diverse 16S-based microbiome studies.

For studies using microbiome data, the ability to robustly combine data from technically and biologically distinct microbiome studies is a crucial means of supporting more robust and clinically relevant inferences. Formidable technical challenges arise when attempting to combine data from technically diverse 16S rRNA gene variable region amplicon sequencing (16S) studies. Closed operational taxonomic units and taxonomy are criticized as being heavily dependent upon reference sets and with limited precision relative to the underlying biology. Phylogenetic placement has been demonstrated to be a promising taxonomy-free manner of harmonizing microbiome data, but it has lacked a validated count-based feature suitable for use in machine learning and association studies. Here we introduce a phylogenetic-placement-based, taxonomy-independent, compositional feature of microbiota: phylotypes. Phylotypes were predictive of clinical outcomes such as obesity or pre-term birth on technically diverse independent validation sets harmonized post hoc. Thus, phylotypes enable the rigorous cross-validation of 16S-based clinical prognostic models and associative microbiome studies.