The Integrative Human Microbiome Project: dynamic analysis of microbiome-host omics profiles during periods of human health and disease

Abstract

Much has been learned about the diversity and distribution of human-associated microbial communities, but we still know little about the biology of the microbiome, how it interacts with the host, and how the host responds to its resident microbiota. The Integrative Human Microbiome Project (iHMP, http://hmp2.org), the second phase of the NIH Human Microbiome Project, will study these interactions by analyzing microbiome and host activities in longitudinal studies of disease-specific cohorts and by creating integrated data sets of microbiome and host functional properties. These data sets will serve as experimental test beds to evaluate new models, methods, and analyses on the interactions of host and microbiome. Here we describe the three models of microbiome-associated human conditions, on the dynamics of preterm birth, inflammatory bowel disease, and type 2 diabetes, and their underlying hypotheses, as well as the multi-omic data types to be collected, integrated, and distributed through public repositories as a community resource.

Figure 1. Integrative Multi-Omic Analysis of the Vaginal and Related Microbiomes in Pregnancy: Sample Collection, Assays, and Data Generation Workflow

Samples from pregnant women and neonates will be collected at clinics associated with Virginia Commonwealth University and the Global Alliance to Prevent Prematurity and Stillbirth (GAPPS). Health questionnaires will be administered and samples collected from multiple body sites over multiple visits throughout pregnancy, at delivery, at discharge, and at follow-up visits. Neonates will be sampled at delivery, discharge, and follow-up visits. A multi-omic approach will probe properties of the host and microbial communities to generate an integrative, longitudinal, and comprehensive data set of 16S rRNA gene surveys, mass spectrometry-based lipidomic profiles, and cytokine assays. A subset of samples will be subjected to whole metagenome and metatranscriptome sequencin for cultivation and isolation of bacterial strains for genome sequencing, and for generation of interactome maps.

Figure 2. Characterizing the Gut Microbial Ecosystem for Diagnosis and Therapy in Inflammatory Bowel Disease: Sample Collection, Assay, and Data Generation Workflow

Samples from Crohn’s disease patients, ulcerative colitis patients, and non-IBD controls are collected at Massachusetts General Hospital (adult new onset), Emory University (pediatric new onset), Cincinnati Children’s Hospital (pediatric new onset), and Cedars-Sinai Medical Center (adult established). From each patient, three different types of samples are collected: longitudinal stool samples, periodic biopsies, and regularly scheduled blood samples. Biopsies are collected as clinically indicated, blood during clinical visits, and stool samples are self-collected by participants at home and shipped directly to a centralized handling and aliquotting pipeline. Multi-omic data generation (primarily, but not entirely, nucleotide sequence- and mass spectroscopy-based) provides microbial, host, and mixed profiles including 16S rRNA gene surveys, whole metagenome and metatranscriptome shotgun sequences, metabolite and protein profiles, single-cell assays, whole virome shotgun sequences, and serological profiles. Each sample is further accompanied by clinical (bloods/biopsies) or self-reported environmental and dietary (stools) metadata.

Figure 3. Microbiome and Host Changes during Respiratory and Other Stress Conditions in Individuals at Risk for Type 2 Diabetes: Sample Collection, Assay, and Data Generation Workflow

All samples are collected at the Stanford Clinical & Translational Research Unit. From each patient in every visit, blood sample and microbiome sample (including nasal swabs and stool and urine samples) are collected. Multi-omic data generation (primarily, but not entirely, nucleotide sequence- and mass spectroscopy-based) will provide profiles of microbial phylogenetic composition, metagenomes, metatranscriptomes, and metaproteomes; host protein profiles, cytokines, and autoantibodies; and global metabolome profiles. Each sample is further accompanied by clinical (blood) or self-reported stress level, environmental, and dietary (stool and urine) metadata.