Dietary plant diversity predicts early life microbiome maturation

Dietary plant diversity predicts early life microbiome maturation | medRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-P4HH5NV'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search Dietary plant diversity predicts early life microbiome maturation View ORCID Profile Teresa K. McDonald , View ORCID Profile Ammara Aqeel , Ben Neubert , Anna Bauer , Sharon Jiang , Olivia Osborne , Nolan Ives , Danting Jiang , Filemon Bucardo , Lester Gutiérrez , Luis Zambrana , Kirsten Jenkins , Jennifer Gilner , Javier Rodriguez , Amanda Lai , Jonathan P. Smith , Rinn Song , Kazi Ahsan , Sheraz Ahmed , Sanam Iram Soomro , Fayaz Umrani , Michael Barratt , Jeffrey I. Gordon , Asad Ali , View ORCID Profile Najeeha Iqbal , View ORCID Profile Jillian H. Hurst , Victoria Martin , Wayne Shreffler , Qian Yuan , View ORCID Profile Joe M. Brown , View ORCID Profile Neeraj K. Surana , Samuel Vilchez , Sylvia Becker-Dreps , View ORCID Profile Lawrence A David doi: https://doi.org/10.1101/2025.02.28.25323117 Teresa K. McDonald 1 Department of Molecular Genetics and Microbiology, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Teresa K. McDonald Ammara Aqeel 1 Department of Molecular Genetics and Microbiology, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ammara Aqeel Ben Neubert 1 Department of Molecular Genetics and Microbiology, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Anna Bauer 1 Department of Molecular Genetics and Microbiology, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Sharon Jiang 1 Department of Molecular Genetics and Microbiology, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Olivia Osborne 1 Department of Molecular Genetics and Microbiology, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Nolan Ives 1 Department of Molecular Genetics and Microbiology, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Danting Jiang 10 Department of Pediatrics, Division of Infectious Diseases, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Filemon Bucardo 6 Department of Family Medicine, University of North Carolina at Chapel Hill , Chapel Hill, North Carolina, USA 8 Department of Microbiology and Immunology, University of North Carolina Chapel Hill , Chapel Hill, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Lester Gutiérrez 23 Centro de Investigación de Enfermedades Tropicales (CIET). Facultad de Microbiología. Universidad de Costa Rica , San José, Costa Rica Find this author on Google Scholar Find this author on PubMed Search for this author on this site Luis Zambrana 7 Department of Epidemiology, University of North Carolina at Chapel Hill , Chapel Hill, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Kirsten Jenkins 10 Department of Pediatrics, Division of Infectious Diseases, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jennifer Gilner 11 Department of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Javier Rodriguez 12 Department of Pediatrics, Children’s Clinical Research Unit, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Amanda Lai 14 The Aquaya Institute , San Anselmo, California, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jonathan P. Smith 16 Department of Health Policy and Management, Yale School of Public Health , New Haven, Connecticut, USA 17 Division of Global HIV and Tuberculosis, U.S. Centers for Disease Control and Prevention , Atlanta, Georgia , USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Rinn Song 18 Oxford Vaccine Group, Department of Pediatrics, University of Oxford , Oxford, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site Kazi Ahsan 19 The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine , St. Louis, MO 63110 USA 20 Newman Center for Gut Microbiome and Nutrition Research, Washington University School of Medicine , St. Louis, MO 63110 USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Sheraz Ahmed 21 Aga Khan University, Medical College , Pakistan Find this author on Google Scholar Find this author on PubMed Search for this author on this site Sanam Iram Soomro 21 Aga Khan University, Medical College , Pakistan Find this author on Google Scholar Find this author on PubMed Search for this author on this site Fayaz Umrani 21 Aga Khan University, Medical College , Pakistan Find this author on Google Scholar Find this author on PubMed Search for this author on this site Michael Barratt 19 The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine , St. Louis, MO 63110 USA 20 Newman Center for Gut Microbiome and Nutrition Research, Washington University School of Medicine , St. Louis, MO 63110 USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jeffrey I. Gordon 19 The Edison Family Center for Genome Sciences and Systems Biology, Washington University School of Medicine , St. Louis, MO 63110 USA 20 Newman Center for Gut Microbiome and Nutrition Research, Washington University School of Medicine , St. Louis, MO 63110 USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Asad Ali 21 Aga Khan University, Medical College , Pakistan Find this author on Google Scholar Find this author on PubMed Search for this author on this site Najeeha Iqbal 21 Aga Khan University, Medical College , Pakistan Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Najeeha Iqbal Jillian H. Hurst 10 Department of Pediatrics, Division of Infectious Diseases, Duke University School of Medicine , Durham, North Carolina, USA 11 Department of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Jillian H. Hurst Victoria Martin 2 Harvard Medical School , Boston, Massachusetts, USA 3 Food Allergy Center, Massachusetts General Hospital , Boston, Massachusetts, USA 4 Division of Pediatric Gastroenterology and Nutrition, Massachusetts General Hospital for Children , Boston, Massachusetts, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Wayne Shreffler 2 Harvard Medical School , Boston, Massachusetts, USA 3 Food Allergy Center, Massachusetts General Hospital , Boston, Massachusetts, USA 5 Division of Pediatric Allergy and Immunology, Massachusetts General Hospital for Children , Boston, Massachusetts, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Qian Yuan 2 Harvard Medical School , Boston, Massachusetts, USA 3 Food Allergy Center, Massachusetts General Hospital , Boston, Massachusetts, USA 4 Division of Pediatric Gastroenterology and Nutrition, Massachusetts General Hospital for Children , Boston, Massachusetts, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Joe M. Brown 13 Department of Environmental Sciences and Engineering, Gillings School of Global Public Health, University of North Carolina at Chapel Hill , Chapel Hill, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Joe M. Brown Neeraj K. Surana 1 Department of Molecular Genetics and Microbiology, Duke University School of Medicine , Durham, North Carolina, USA 22 Department of Pediatrics, Duke University School of Medicine , Durham, North Carolina, USA 15 Duke Microbiome Center, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Neeraj K. Surana Samuel Vilchez 6 Department of Family Medicine, University of North Carolina at Chapel Hill , Chapel Hill, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Sylvia Becker-Dreps 6 Department of Family Medicine, University of North Carolina at Chapel Hill , Chapel Hill, North Carolina, USA 7 Department of Epidemiology, University of North Carolina at Chapel Hill , Chapel Hill, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Lawrence A David 1 Department of Molecular Genetics and Microbiology, Duke University School of Medicine , Durham, North Carolina, USA 15 Duke Microbiome Center, Duke University School of Medicine , Durham, North Carolina, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Lawrence A David For correspondence: lawrence.david{at}duke.edu Abstract Full Text Info/History Metrics Data/Code Preview PDF Summary Despite established links between the infant gut microbiome and health, how complementary feeding shapes colonization remains unclear. Using FoodSeq, a DNA-based dietary assessment technique, we surveyed food intake across 729 children (0-3 y) from North and Central America, Africa, and Asia. We detected 199 unique plant food sequences, with only eight staples shared across all countries. Despite this global variation, early-life diets followed a common trajectory: weaning stage emerged as the dominant dietary signature across populations. Crucially, dietary diversity did not correlate with gut microbial diversity. Instead, dietary diversity and weaning stage specifically predicted the abundance of adult-associated bacteria. Our findings support a two-stage maturation model: an initial milk-dominated phase, followed by a diet-responsive phase that yields adult-like communities. Monitoring and promoting plant dietary diversity may thus support timely microbiome maturation worldwide. Introduction In the first years of life, the human gut transitions from a sterile state to a diverse microbial ecosystem. During this transition, the gut microbiome transforms from an ‘immature’ state -- characterized by a limited set of milk-adapted taxa -- to a ‘mature’ adult-like community harboring fiber-degrading taxa and a diversity of other commensal species 1 – 3 . Proper microbial succession is fundamental to healthy metabolism, immune development, and disease resistance. Indeed, disruptions in early life colonization and microbiome maturation increase risk of conditions such as obesity 4 , allergy 5 , and Type I diabetes 6 . Consequently, the first years of life represent a critical window when microbiome development can have lifelong health effects. Despite the importance of the gut microbiome in healthy infant development, the specific factors guiding early microbiome colonization remain less clear. Well-established influences include antibiotic exposure, which reduces bacterial diversity and stability; caesarean delivery, which alters the initial microbial repertoire; and formula feeding, which lowers Bifidobacterium abundance yet increases overall microbial diversity 7 . Another pivotal factor is cessation of breastfeeding 3 , 2 , 8 , which triggers a decline in the abundance of milk-adapted microbial taxa. One area of microbiome development, that remains underdeveloped, however, is the precise role that complementary feeding plays in microbial succession 1 . While diet is known to reproducibly alter the gut microbiome in adults 9 , few studies have found evidence directly linking specific dietary factors to microbiome maturation by concurrently measuring diet and the microbiome in young children. Exceptions have either tested narrow food interventions 10 – 13 or were conducted within a single cultural and dietary context 14 – 17 . Other studies have measured microbiome composition across more diverse dietary traditions 8 , 18 ; still such studies have been limited in sample size or have often lacked detailed dietary data, leading comparisons to be on population-level dietary patterns rather than individual-level diet metrics. A major barrier to understanding the relationship between diet and microbiome maturation is the challenge of accurately characterizing infant diet. Surveys, the gold standard and traditional approach for collecting dietary data, are labor-intensive 56 require baseline levels of literacy 51 , incompletely capture minority or ethnic foods 52 – 54 , and are vulnerable to social desirability bias 19 , 20 . Dietary assessment in children poses additional obstacles, as it relies on secondhand reporting by parents and guardians who may not observe all meals when children are under the supervision of other caregivers. In the United States, for instance, a quarter of 2-5 year olds’ energy intake is consumed outside the home 21 . Such incomplete observation contributes to reporting bias exemplified by validation studies with doubly-labeled water that show infant dietary intake is overestimated by 7-29% 22 – 27 . Finally, the technical challenges with infant dietary assessment only multiply in multi-country settings, as assessments must be adapted for each different geographic, cultural, and linguistic setting 28 For example, there are over 100 versions of a standardized survey for comparing population-level global diets 29 . In concert, these barriers have prevented systematic investigation of the relationship between early life microbiome development and infant diet across diverse human populations. Remarkable diversity in early-life diets worldwide To perform a global survey of infant microbiome development and diet, we assembled a meta-cohort from existing longitudinal studies and population health surveillance surveys of 1036 fecal samples from 729 children aged 0-3 years from five countries across multiple continents: the USA (n=144), Nicaragua (n=457), Cambodia (n=178), Kenya (n=228), and Pakistan (n=29). These sites were selected to encompass wide variation in geography, socioeconomic status, and dietary traditions, as well as varying burdens of infectious diseases common in early childhood. Stool samples and dietary information were collected; further details on cohort characteristics and methodologies are provided in the methods and Table S1 . View this table: View inline View popup Download powerpoint Table S1: Cohort Characteristics We applied FoodSeq, an objective DNA sequencing-based dietary assessment 30 , to our cohort. FoodSeq is an amplicon sequencing-based technique that sequences the leucine gene ( trnL ) from plant plastids or the 12S rRNA gene from animal mitochondria. By mapping trnL and 12S sequencing reads to a curated reference database, FoodSeq identifies and quantifies foods of plant and animal origins at a taxonomic level. Unlike traditional dietary surveys, FoodSeq provides objective measurements that are independent of caregiver reporting and remain consistent across varying income levels, health conditions, and infant feeding practices. The diet data generated by FoodSeq also has the advantage of being temporally synchronized with corresponding microbiome data, as both sets of genomic analyses can be generated from the same fecal sample. FoodSeq revealed extensive variety and breadth in early life diets across our global cohort ( Fig. 1 ). With plant FoodSeq, we detected 113 species and 86 assigned sequence variants (ASVs) that matched at a higher level (e.g. genus), encompassing 389 potential dietary plant species. Of the plant ASVs, 42% were detected exclusively in a single country ( Fig. S1A ), with unique country-specific foods ranging from longan and lotus root in Cambodia, to jute mallow and cassava in Kenya, to raspberry in the USA ( Fig. 1A ). By contrast, just eight global staples 31 showed consistent prevalence in all five countries (detected in >5% of each country’s samples): wheat, corn, rice, banana/plantain, tomatoes and other nightshades (potato, etc.), mangifera (mangos), and alliums (onion, leek, garlic, chives) ( Fig. 1A , foods in bold). Download figure Open in new tab Fig. S1 (A) Distribution of foods by how many countries they were each detected in. A food was counted as ‘detected’ if it was detected in at least 1 sample in that country (>0% threshold), or at least 1% of samples in that country (>1% threshold). (B) Animal FoodSeq diversity by country. Download figure Open in new tab Fig. 1: Early life diet is diverse and varies globally. (A) Dendrogram of food prevalences. Rare foods (<1% frequency in at least one country) were filtered out prior to plotting. Sequences that matched to a higher taxonomic level than species are labeled with the lowest taxonomic level assigned (plants) or a representative taxonomic level (animals), and then a common name of a food that sequence matches. Sequences with multiple foods or food names associated with them are labeled with ‘etc’. The 8 plant and 2 animal foods detected in at least 5% of samples in all 5 countries are labeled in bold. (B) Top 5 most prevalent foods in each country. Despite the breadth of plant species in our global survey, principal component analysis (PCA) revealed that the primary axis of dietary variation, PC1, was likely driven by the overall presence of plant matter in the diet rather than specific dietary composition ( Fig. 2A-B ). Unlike subsequent PCs, which displayed mixed positive and negative loadings indicative of tradeoffs in plant species abundance, PC1 exhibited exclusively positive loadings for common foods ( Fig. 2C ). This pattern suggests that PC1 captures the extent of plant consumption rather than regional dietary differences. Supporting this interpretation, PC1 was strongly correlated with both infant age ( Fig. 2D , Spearman’s π =0.25, p<0.001) and overall plant FoodSeq richness (pFR, the number of unique sequences assigned to our human foods database in each sample; Fig. 2E , π =0.69, p<0.001). These associations are consistent with the expected trajectory of weaning as infants gradually incorporate more variety of solid foods into their diets as they mature. Direct comparisons of dietary diversity and age confirmed our interpretation, showing that pFR consistently increased across the first year of life until stabilizing ( Fig. 2G ; Spearman’s π =0.27, p<0.001). Taken together, these findings argue that FoodSeq effectively captures the dietary shifts associated with weaning. Download figure Open in new tab Fig. 2: Dietary variation across age and country. (A) Plant FoodSeq PCA colored by country. Country accounted for 15% of variation, age accounted for 0.6% variation, and dietary diversity accounted for 5% of variation. (PERMANOVA, p-value <0.001 for all). (B) PCA colored by age (C) PCA loadings. (D) Correlation between age and PC1. (E) Correlation between pFR and PC1. (G) pFR over time. A logistic line (black) was fit to the data. The red dashed line represents the plateau point at which the increase in pFR decelerates, calculated as the x value at which the first derivative of the line reaches 50% of its maximum. (H) pFR by country. While weaning stage dominated the first PC, the second PC revealed country-specific dietary signatures partitioned by their relative intakes of regional staple plants like rice (enriched in Cambodia), banana/plantain (enriched in Nicaragua), or sorghum and millet (enriched in Kenya). Cross-country differences also emerged in dietary diversity more broadly, with median pFR ranging from 4 in Nicaragua to 10 in Cambodia ( Fig. 2H ). Such values aligned with previously reported food variety scores, which found mean food variety ranges from 5.5 to 10 among children from 6-24 months in Africa and Central America 32 – 34 . The timing and rate of dietary diversification also varied by country. For instance, the diets of Cambodian children diversified early and rapidly until pFR plateaued at 8 months of age, whereas US infants showed a more gradual rise in pFR until 13 months ( Fig. 3B ). Download figure Open in new tab Fig. 3 The early life microbiome varies primarily by geography and age. (A) Microbiome diversity increases with age. A logistic line (black) was fit to the data. The red dashed line represents the point at which diversity plateaus with respect to age, defined as the point at which the first derivative of the line reached 50% of its maximum. (B) Dietary diversity plateaus prior to microbiome diversity. A logistic line (black) was fit to the data and the plateau point (red dashed line) calculated as in (A). (C) PCA of 16S rRNA sequencing data colored by country. (D) PCA of 16S data colored by age in months. We note that in contrast to the diversity of plant-associated species and dietary patterns observed in our cohort, we detected a more limited breadth of dietary animals. A total of 28 dietary animal species were detected. Observed animals included widely utilized livestock such as pig, chicken, and cow, as well as more region-specific animals including water buffalo, the primary source of milk in Pakistan, and various fish species in Cambodia. Overall, however, 41% of infant samples had no non-human animal DNA detected, and only 33% had more than one animal species detected. Only the US and Cambodia had a non-zero median count of animal species present ( Fig. S1B ). The limited detection of dietary animals was consistent with the higher cost of animal-based foods and the majority of human animal-based foods originating from cattle, poultry, and pigs 28 for the locations of infants sampled here. Given the limited detected range of animal species, the distinct nutritional characteristics of different animal products from the same species (beef, dairy), as well as the known importance of nutrients like fiber for the gut microbiome, we focused our subsequent analyses on plant-based foods and gut microbiome development. Patterns of microbiome maturation Prior to analyzing the relationship between early-life dietary and microbiome development, we first confirmed that our cohort followed well-established patterns of early-life gut microbiome maturation 3 . Microbiome alpha-diversity increased steadily from birth ( Fig. 3A ) through age two, regardless of country ( Fig. 3B ). In terms of inter-individual variation (beta-diversity), both country of origin ( Fig. 3C ) and age ( Fig. 3D ) were significant factors (country R 2 =0.4675, p=0.001, age R 2 =0.4721, p=0.001; EnvFit on Unweighted Unifrac. Fig. S2A ). In cohorts where data were available on breastfeeding status (USA, Nicaragua, Cambodia) and birth mode (USA, Nicaragua), both factors were associated with microbiome composition. (Breastfeeding: R 2 = 0.1377, p=0.001; Birth mode: R 2 = 0.0088, p=0.014, Envfit on unweighted Unifrac, controlling for age and country). Thus, consistent with previous literature, we observed both increased diversification of the microbiome with age, as well as notable roles for breastfeeding status and birth mode in shaping microbiome structure in our cohort 35 . Download figure Open in new tab Fig. S2 (A) Unweighted Unifrac ordination. (B) Plant and microbiome diversity versus age, with Pakistan and Nicaragua included (see: Fig. 3B ). (C) Scatterplots showing relationships between microbial Shannon diversity and pFR within country. (D) Scatterplot showing relationship between microbial Shannon diversity and pFR. (E) Scatterplot showing relationships between microbial richness and pFR. For C-E, line plots the coefficient of pFR derived from a generalized linear model controlling for age (C-E), and, where applicable, country (D-E) and breastfeeding status (C, Cambodia and Nicaragua). The specific bacterial taxa observed in our infant cohort also aligned with prior studies 1 – 3 , 8 , as hierarchical clustering revealed a microbial succession pattern ( Fig. S3A ). We identified an early-life cluster (Cluster 9), enriched in Bifidobacterium and Streptococcus, and a ‘transitional’ cluster (Cluster 1), abundant at ages 12-18 months, characterized by plant degraders, Blautia and Ruminococcus, as well as the probiotic genus Ligilactobacillus ( Fig. 4A ). Download figure Open in new tab Fig. S3: (a) Heatmap plotting age-differential microbiome clusters. (b) Late microbiome abundance correlated with pFR within country while controlling for breastfeeding status, where available, and age. Similar to microbiome diversity, late microbiome abundance is positively correlated with plant richness in Kenya and the USA. Download figure Open in new tab Fig. 4 Plant dietary diversity is correlated with the abundance of mature-microbiome marker bacteria. (A) Cluster abundance over time. (B) The abundance of the ‘Early Microbiome’ cluster, plotted against plant richness, (C) The abundance of the ‘Transitional Microbiome’ cluster, plotted against plant richness. ( D-E) The abundance of the ‘Late Microbiome’ cluster, plotted against plant richness (D) and PC1 (E). For B-E, lines plot the coefficient derived from a generalized linear model (GLM) controlling for age and country. Following this transition, a late-microbiome cluster (Cluster 5) emerged between 21-36 months that more closely resembled the adult microbiome, featuring Bacteroides vulgatus and Faecalibacterium prausnitzii, which ferment plant fiber into butyrate and other short chain fatty acids, and Prevotella copri, a core fiber-metabolizing taxon 36 ( Fig. 4A ). In parallel, we constructed a Random Forest model that successfully predicted infant age given input microbiome data. The RF model identified Faecalibacterium and Bifidobacterium as top predictor variables, mirroring both our hierarchical clustering results and previously published models of infant microbiome development 1 (Fig.S4). Dietary diversity does not predict microbiome diversity Having found that infant microbiota in our cohort matured as expected from previous studies, we next compared patterns of dietary maturation to gut microbiome maturation. Surprisingly, we did not see consistent evidence that dietary diversity predicted gut microbial diversity. Microbial alpha diversity continued to increase until at least 14-16 months of age ( Fig. 3A ) well after dietary diversity (as measured by pFR) had plateaued ( Fig. 3B ; 2G). We did observe associations between dietary diversity and microbiome diversity within several countries but associations were not consistent; for example, pFR correlated with microbiome Shannon diversity in Kenya and the United States, whereas in Nicaragua, the relationship was reversed ( Fig. S2C ). Further, when we constructed global models controlling for age and country, pFR did not correlate wtih overall microbiome diversity measures (Shannon diversity and 16S richness both non-significant ( Fig. S2D , F). Thus, although children began eating complex diets soon after introduction of solid foods, the gut microbiome continued to diversify even after dietary complexity had been reached. Diet is associated with later stage microbes during gut colonization We next hypothesized that although dietary diversity may not directly correlate with overall microbial diversification, plant richness and weaning stage would be associated with microbiome maturity. Indeed, despite substantial geographic variation in diet, we found pFR was positively correlated with the ‘Transitional Microbiome’ cluster, comprised of Blautia and Ruminococcus , and the ‘Late Microbiome’ cluster, comprised of Bacteroides , Faecalibacterium , and Prevotella , even after controlling for age and country ( Fig. 4C, D ). Likewise, PC1 was positively correlated with ’Late Microbiome’ cluster abundance ( Fig. 4E ). Given that we sourced fecal samples across multiple study sites, we re-fitted all models while (i) adjusting for diarrhea, weight-for-height z-score, RUTF intake, and sequencing depth, and (ii) adding participation in the Cambodian nutrition-education intervention. None of these adjustments altered the direction or significance of the associations between plant richness (pFR) and the Transitional or Late microbiome clusters, nor did they create any association with Shannon diversity ( Fig. S5 ,6). Download figure Open in new tab Fig. S4 (A) Model fit for classifier predicting age from shuffled or true 16S rRNA relative abundance data. (B) Top predictors from Random Forest model. (C) Exact sequence matches between the top 100 predictors in the Random Forest model, the taxa identified in the Early, Transitional, and Late clusters, and a previous study that first employed this analysis. (D) Plot of microbiome age versus actual age. Red dashed line is a theoretical perfect concordance between the two. Download figure Open in new tab Fig. S5 Generalized linear models controlling for country-specific variables. (A) Models controlling for RUTF consumption. (B) Models controlling for weight-for-height z score (WHZ). (C) Models controlling for study intervention group within the Cambodian cohort. (D) Models controlling for whether or not Cambodian children were in an intervention group that included a nutrition intervention. (E) Models controlling for diarrhea. Each column represents an outcome variable, and each row represents a predictor variable. Results are grouped by whether the test was run on all samples with non-NA values for that variable (‘Global’) or within a specific country. Missing countries did not collect data for that variable. Download figure Open in new tab Fig. S6. Summary of generalized linear model results for other diet metrics. (A) Model results for pFR calculated from data rarefied to 500 reads. (B) Model results for animal dietary diversity (aFR). Y axis is the dataset the test was run on (‘global’ for all samples, or within a specific country.) Pakistan was omitted due to low sample size (n=29). X axis represents the coefficient, indicative of effect size, of that variable in the model. Ultimately, as increased dietary diversity (i.e., pFR) and weaning stage (i.e., PC1) both reflect complementary feeding and the introduction of an adult-like diet, our findings support the hypothesis that introduction of an adult-like diet stimulates maturation toward an adult-like microbiome. Notably, the ’Early Microbiome’ cluster did not correlate with dietary diversity ( Fig. 4B ) or PC1, consistent with the hypothesis that milk consumption, rather than solid food introduction, primarily shapes gut microbial ecology prior to weaning. These results suggest that the transition to an adult-like microbiome specifically is driven by the cumulative introduction of diverse solid foods, but that other transitions within the microbiome maturation process, such as overall microbiome diversification, are not driven by solid food diversity. These transitions are furthermore unlikely to be driven by consumption of particular foods or specific dietary patterns, as we observed conserved microbiome microbial responses to diversity across countries with highly unique and non-overlapping diets. More broadly, our findings point to a developmental sequence wherein milk consumption governs early microbiome composition, while the subsequent introduction of solid plant foodsdrives the transition toward a mature gut microbial community. Discussion In concert, our findings here refine existing models linking weaning to microbiome development in the infant gut. Our data do not reveal simple linear relationships between dietary and microbiome diversity across early life. This negative finding, however, is consistent with our increasing understanding that microbial ecosystem diversity does not necessarily predict community function 37 , 38 . Instead, our data support a two-stage developmental model: an early phase governed by milk consumption, followed by a maturation phase shaped by dietary diversity. In this second stage, plant dietary diversity and the child’s physiologic age together predict the colonization by select groups of gut bacterial taxa associated with adult-like microbiome function. Notably, despite the broad range of complementary feeding patterns across our geographically diverse cohorts, we observed similar successional trends, underscoring the functional redundancy 39 that gut microbes can share, regardless of environmental differences. Early life nutrition remains one of the most critical challenges in global health, with lasting consequences for child development and lifelong health outcomes 40 . Our findings confirm that adequate and diverse plant consumption during the complementary feeding period foster maturation of the gut microbiome toward an adult-like state rich in fiber-degrading bacteria, extending the known benefits of dietary diversity beyond micronutrient adequacy 41 , 42 and food allergy prevention 43 . Importantly, we show that simply counting the number of distinct plant foods effectively measures this microbial transition, potentially offering a more accessible and practical complement to widely used dietary indices such as the WHO minimum dietary diversity score (MDD), which assesses diversity as consumption across at least five of eight food groups. 44 Implementing straightforward, ‘pro-gut’ complementary feeding guidance based on overall plant dietary richness could enable easier and more precise monitoring by caregivers and health workers across diverse settings, ultimately supporting appropriate microbiome maturation among infants worldwide. Materials and Methods Study populations and stool sampling Cambodia Cambodian samples were obtained as extracted DNA from stool samples collected as part of the Integrated Nutrition, Hygiene and Sanitation Project (NOURISH), conducted across three rural provinces 45 . The NOURISH study involved both nutrition and sanitation interventions. The nutrition intervention specifically consisted of caregiver nutrition education, direct cash payments contingent upon regular health visit attendance, and provision of two $5 food vouchers. Of the Cambodian samples in our study, 101 (56.7%) originated from children enrolled in either of the nutrition intervention arms (arm 1: nutrition intervention only; arm 2: nutrition intervention+sanitation). Our selection prioritized samples from children who had not experienced diarrhea or fever within the week preceding sample collection; however, 14 samples came from children who had recently experienced diarrhea, 156 from children without recent diarrhea, and five lacking diarrhea data. Additionally, seven samples came from children who had received RUTF or a local alternative, independent of the intervention protocol. The cohort overall exhibited a high burden of enteric pathogens, although most pathogen-positive children were asymptomatic at the time of sample collection. Kenya Kenyan samples were obtained as extracted DNA from stool samples collected in Kisumu, Kenya as part of the TOTO study 46 . The TOTO cohort comprised children generally representative of the broader Kenyan pediatric population. Children were enrolled in the TOTO study when presenting for clinical care with symptoms of respiratory illness raising clinical suspicion for tuberculosis infection. Full study enrollment criteria and clinical characteristics of the original TOTO study have been previously published 46 . Of the 228 Kenyan samples included in our analysis, 70 (30.8%) reported receiving nutritional supplmentation with RUTF or a local alternative as part of clinical care, independent of microbiome sampling protocol. Nicaragua Nicaraguan samples were obtained as stool slurries (stored 1:1 in phosphate-buffered saline, PBS) from the Sapovirus -Associated Gastro-Enteritis (SAGE) study 47 , which enrolled children at risk of diarrheal disease in León, Nicarauga. Full enrollment criteria and clinical characteristics of the original SAGE study have been previously published 47 . In addition to the previously described control (n=33) and diarrheal (n=32) samples, we obtained n=391 additional stool samples collected for monthly surveillance from study participants. Non-diarrheal surveillance samples were specifically selected from children who had not experienced a diarrhea episode within the preceeding month. Only surveillance samples from children aged 2-12 months were included in our 16S microbiome analysis. Pakistan The Pakistan samples were obtained as extracted DNA from fecal samples collected as part of the Study of Environmental Enteropathy (SEEM), a prospective longitudinal cohort study, conducted in Matiari, a rural area in Sindh, Pakistan, between March 2016 and March 2019. The SEEM study collected samples from 50 healthy control children and longitudinal samples from malnourished children (WHZ 0), none of whom received nutritional interventions. Approximately 1g of stool was collected from each participant by community health workers within 10 minutes of defecation. Samples were immediately transported from field sites in 4°C cooler boxes, stored at -80°C upon arrival at the laboratory, and shipped on dry ice for subsequent analysis. United States US samples consisted of frozen solid stool or stool slurry obtained from participants enrolled in the Project HOPE 1000 study 48 (Durham, NC, USA) and Gastrointestinal Microbiome & Allergic Proctocolitis Study (GMAP) study (Boston, MA, USA) 49 . Project HOPE 1000 aims to develop a biorepository of longitudinal samples collected from healthy infants over the first 1000 days of life, with the goal of identifying early life determinants of health. The GMAP study prospectively collected stool samples from children, some of whom subsequently developed allergies and/or proctocolitis; however, GMAP samples analyzed in this study were exclusively from infants who did not develop allergic disease. Breastfeeding status was documented at the time of each sample collection for Project HOPE 1000. For GMAP, breastfeeding status was recorded at birth and 4 months of age, but not at the exact time of stool sample collection. Therefore, infants exclusively receiving formula at 4 months of age in GMAP were presumed to not be breastfed at the time of stool collection. DNA Extraction Cambodia: Stool samples were preserved 1:1 in Zymo DNA/RNA Shield buffer (Zymo Research, Irvine, CA) and then stored at -20°C. They were then extracted using the QIAamp 96 Virus QIAcube HT (Qiagen, Germany) protocol. 200mg of solid stool or 200uL of liquid stool were combined with 1mL of stool lysis buffer (ASL, Qiagen, Germany) and vortexed in an SK38 soil grinding tube (Bertin Corp., Rockville, MD) for five minutes (Vortex Genie 2, Scientific Industries, Bohemia, NY). Following a ten minute incubation at room temperature, the lysate was centrifuged at 12,000 g for two minutes (Thermo Fisher Scientific, Waltham, MA). 200uL of this lysate was then used for DNA/RNA extraction with the QIAamp 96 Virus QIAcube HT protocol. Kenya: 1g of stool was diluted in 10mL of sterile water and emulsified with sterile glass beads. The stool slurry was then filtered. DNA was extracted using the MagAttract PowerSoil DNA EP kit (Qiagen, Germany) according to the manufacturer’s instructions. Pakistan : 50mg of frozen fecal sample was extracted using phenol-chloroform extraction. Briefly, samples were ground in liquid nitrogen using a mortar and pestle. 50mg of this was then transferred to a 2mL vial. 500uL of 0.1mm diameter zirconia/silica beads were added to each sample, along with 500 µL of phenol: chloroform: isoamyl alcohol (25:24:1), 210 µL of 20% SDS and 500 µL of buffer (200 mM NaCl, 200 mM Trizma base, 20 mM EDTA). Samples underwent bead beating (Mini-Beadbeater-8; Biospec) to extract DNA. Extracted DNA was purified, (Qiaquick columns, Qiagen), then eluted in 70 µL of Tris-EDTA (TE) buffer, and then quantified (Quant-iT dsDNA broad range kit; Invitrogen). Each DNA sample was adjusted to a concentration of 1 ng/uL. Nicaragua and USA The Nicaraguan and USA (Durham, Boston) stool samples were diluted 1:10 in filtered phosphate buffered saline (0.2uM filter) and then stored at -80°C. Samples were thawed on ice in an anaerobic chamber. Samples were vortexed 30s and then 250uL aliquots were taken for DNA extraction. Slurry aliquots were then randomized and extracted using the Power Soil Pro MagAttract kit (Qiagen, Germany) according to manufacturer’s instructions. DNA Sequencing 16S rRNA Sequencing All samples were PCR-amplified using primers targeted to the V4 region of the 16S rRNA gene using primers 515F and 806R as previously described. 50 Kenyan samples were quantified using a Qubit dsDNA HS assay kit (ThermoFisher), pooled in equimolar concentration, and sequenced on a MiSeq sequencer (Illumina; 2x250 base pair). Cambodian, Nicaraguan, Pakistan, and USA samples were processed as previously described 50 – 52 , diluted to a 10nM concentration, and sequenced using an Illumina MiniSeq and a MiniSeq High Output kit (FC420-1004) with paired-end 150-bp reads. FoodSeq FoodSeq was carried out as described in previously published protocols. 53 For plant FoodSeq, samples were amplified in a 10uL volume with 3 μL of extracted DNA template and either a) 0.5 μL each of 10 μM forward and reverse primers (IDT), 2 μL of 5X KAPA HiFi buffer, 0.3 μL of 10 mM dNTPs, 0.1 μL of 100X SYBR Green I (Life Technologies), 0.1 μL KAPA HiFi polymerase, 3.5 μL nuclease-free water, or b) 5uL of KAPA HiFi 2X, 0.3uL each of 10 μM forward and reverse primers (IDT), 0.1uL of 100X SYBR Green I (Life Technologies) and 1.3uL of nuclease-free water. Method (b) uses the same components as method (a), but in a premixed format with a slightly higher concentration of polymerase. Cycling conditions were 95°C for 3 minutes, followed by 35 cycles of 98°C for 20 seconds, 63°C for 15 seconds, and 72°C for 15 seconds. For animal FoodSeq, samples were amplified in a 10uL volume with 1uL of extracted DNA template and 0.5uL of 10uM forward and reverse primer and 1uL of 100uM human blocking primer (IDT), 0.1 uL of 100X SYBR Green, 0.25uL of 20mg/mL bovine serum albumin, and 5uL of 2X Accustart. The amplification primers were 12SV5F and 12SV5R 54 with Illumina overhang adapter sequences added at the 5′ end, and the human blocking primer was DeBarba14 HomoB 55 . Cycling conditions were an initial denaturation at 94°C for 3 minutes, followed by 35 cycles of 94°C for 20 seconds, 57°C for 15 seconds, and 72°C for 1 minute. For both plant and animal FoodSeq, primary amplification was followed by secondary PCR amplification to add Illumina adapters and dual 8-bp indices for sample multiplexing. Samples were amplified in a 50uL volume using 5uL primary PCR product diluted 1:100 in nuclease free water, 25uL KAPA HiFi 2X Buffer, 0.5uL 100X SYBR Green I, 9.5uL nuclease free water, 5uL each of 2.5uM forward and reverse indexing primers. Amplicons were cleaned (Ampure XP, Beckman Coulter, Brea, CA), quantified (QuantIT dsDNA assay kit, Invitrogen, Waltham, MA), and pooled in equimolar ratios to create a sequencing pool. If samples were too low in concentration to be pooled at a equimolar ratio, they were added up to a set volume of 15-20uL. The pooled library was then concentrated, gel purified, quantified by both fluorimeter and either qPCR or BioAnalyzer (Agilent, Santa Clara, CA). If trnL and 12SV5 amplicons were to be sequenced on the same run, they were combined at this point in equimolar amounts. Final sequencing library was spiked with 30% PhiX (Illumina, San Diego, CA) to mitigate low nucleotide diversity. Libraries were sequenced on an Illumina MiniSeq according to manufacturer’s instructions using a 150-cycle Mid or High kit, (Illumina, San Diego, CA, USA), depending on the number of samples in each pool. Data analysis All packages used are available as a Singularity container hosted by the Duke Gitlabs ( https://gitlab.oit.duke.edu/lad-lab ). 16S rRNA All code used in bioinformatic pipeline to process 16S rRNA amplicon data can be found on Github at tjk30/16S-pipeline. To summarize, following conversion to fastq format using Illumina’s bcl2fastq package, DNA sequencing results were demultiplexed using qiime2 with the --p-no-golay-error-correction flag. Trimmomatic was used to remove primer sequences. Dada2 (v1.22.0) was then used to filter, trim, merge, and assign species to reads using the SILVA database v138.1. Samples with 0. Shannon diversity was calculated using estimate_richness() from the phyloseq package (v.1.48.0) in R (v.4.1). Unweighted Unifrac was performed using the SILVA v138.1 SSU phylogenetic tree and the Unifrac() function from the phyloseq package. Ordination of sample distances was then performed using cmdscale() from the R package stats (v4.4.1) with the flag eig=TRUE. Envfit was run using the envfit() function from the R package vegan ( v2.6.6.1). For calculation of the variation captured by age and country, both variables were included in the model. For the breastfeeding and birth mode models, each model was run controlling for country and age. For microbiome clustering analysis, we excluded samples from children known to be experiencing diarrhea at the time of stool sample collection (n=33/934) or acute malnutrition (WHZ <-2), as these conditions are known to significantly impact gut microbiome composition 56 , 1 . Diarrhea data were available only from the Cambodian and Nicaraguan cohorts. Age was binned as follows: 0-2mo, 3-5mo, 6-8mo, 8-10mo, 11-13mo, 14-16mo,17-19mo, 20-22mo, 23-25mo, and 26+mo (binned together due to limited sampling >24mo in Cambodia and USA). Clustering was limited to 0-36 month old children to ensure that each age bin consisted of children from multiple countries, as only Kenya provided samples from children older than 36 months. Taxa were filtered to those with at least 0.5% relative abundance in at least one age bin. Rare taxa, defined as taxa detected in two or fewer samples, were removed from analyses. The samples that passed filtering criteria (n=818) were then used to calculate the average relative abundance of each ASV within each age bin. In order to normalize across taxa with different relative abundance ranges across age, relative abundances for each taxon were converted to z-scores. Clustering analysis was performed on the Euclidean distances of the taxa z-scores using Unweighted Pair Group Method with Arithmetic Mean (UPGMA; implemented with the hclust() function from the R package stats , method=’average’). Clustering analysis was then used to generate an k=10 clusters using the stats function cutree(). After identifying microbiome clusters in the subset of healthy children, we reincorporated the excluded "sick" samples back into the dataset for subsequent analyses, as we wanted to comprehensively evaluate the identified microbial clusters within the context of broader childhood conditions and maximize statistical power. We calculated the cluster abundances for Clusters 1-10 for all samples in the full dataset (n=934) by merging the taxa identified within each cluster and then performing centered log ratio (clr) normalization using the R package microbiome (v1.26.0). FoodSeq All code and databases used in the bioinformatic pipeline to process FoodSeq amplicon data can be found on Github at LAD-lab/mb-pipeline. Bioinformatic processing was conducted as previously described 30 . To summarize, following conversion to fastq format and demultiplexing using Illumina’s bcl2fastq package, primer sequences were trimmed using the qiime2 cutadapt wrapper with the following flags: --p-error-rate 0.15, --p-minimum-length 1, --p-overlap 5, p-discard-untrimmed. Paired reads were then merged using the qiime2 dada2 wrapper with the following flags: --p-trunc-len-f 0, --p-trunc-len-r 0, --p-max-ee-f 2, --p-max-ee-r 2, --p-trunc-q 2, p-min-overlap 12, --p-pooling-method ’independent’. As we used a non-error-correcting polymerase for the 12SV5 PCR amplification reactions to maximize human blocking primer efficiency (resulting in more error-prone amplification), ASVs mapping to the same species were glommed using tax_glom() in the R package phyloseq prior to analysis. This meant that ASVs were assigned to the most similar species in the database used from Schneider et al. 57 pFR was calculated as the sum of unique assigned ASVs that had any detectable reads. Unassigned ASVs that did not match a human food in our database were excluded from pFR. Samples with a pFR >5 standard deviations outside the median were excluded from further analysis. For food clustering analysis, foods that did not have at least one read in one sample were removed and a matrix was generated of the percent of samples in each country where each assigned ASV was detected. Clustering was performed on the Euclidean distance between each food based on its prevalence in the 5 countries using the “complete” method in the function hclust() from the R package stats . The R package ggtree was then used for dendrogram visualization. Statistical Analysis Data Ordinations To account for the compositional nature of the data, we calculated the centered-log ratio (CLR) transformation of DNA sequencing read counts using the transform() function of microbiome v1.26.0. Principal component analysis (PCA) was then performed on CLR-transformed read counts using the prcomp() function in the stats R package (v4.4.1). Linear model design Generalized linear models were used to model the relationship between dietary factors and microbiome outcomes. We included age as a covariate in our models due to the strong relationships between age and our microbiome outcomes of interest established in previous literature. 1 We further controlled for country due to differences in age distribution, malnutrition, enteric pathogen exposure, and breastfeeding duration between the countries assessed. Our choice to include interaction terms in our models was determined by choosing the model with the lowest AIC and significant chi-squared p-value in a likelihood ratio test between the reduced model and saturated model. The relationship between dietary diversity and microbial diversity was modeled using the call glm(microbiome_diversity ∼ pFR + country + age_months + country:age_months + pFR:country, family = ’gaussian’) . The relationship between dietary PC1 and microbiome clusters was modeled using the call glm(cluster ∼ age_months + country + country:age_months + PC1 + country:PC1 + age_months:PC1, family = ’gaussian’) . Cluster associations with subsequent PCs were not tested, as country was highly collinear with PC2, accounting for half of its variance (ANOVA, p < 2e–16; R² ≈ 0.50 for country), which would render PC2 effects indistinguishable from country-level associations. When fitting within-country models, the following models were used, depending on available covariate data: USA and Kenya (breastfeeding data not available): glm(microbiome_outcome ∼ diet_variable + age_months, family = ’gaussian’) Pakistan: (lacked breastfeeding data; all samples came from one timepoint (12-15 months): glm(microbiome_outcome ∼ diet_variable, family = ’gaussian’) Cambodia and Nicaragua (breastfeeding data available): glm(microbiome_outcome ∼ diet_variable + currently_breastfed + age_months + currently_breastfed:age_months + diet_variable:currently_breastfed, family= ’gaussian’) When accounting for the effects of WHZ on microbiome richness, the following models were used: Kenya: glm(microbiome_diversity ∼ WHZ + pFR + age_months + sex) Cambodia: glm(microbiome_diversity ∼ WHZ + pFR + age_months + sex + currently_breastfed + pFR:currently_breastfed + currently_breastfed:age_months + WHZ:currently_breastfed) Institutional and Ethical Review Stool and extracted stool DNA samples were obtained through collaborations with existing studies via an IRB exemption (Pro00100567) from the Instutitional Review Board at Duke University. Use of Kenyan pediatric stool samples in this analysis was reviewed and approved by the institutional review boards of the US Centers for Disease Control and Prevention, the Kenya Medical Research Institute, and the Jaramogi Oginga Odinga Hospital. Written informed consent for the use of these stool samples was given by parents or legal guardians of participants. U.S. CDC Disclaimer This work has been supported in part by the President’s Emergency Plan for AIDS Relief through the U.S. Centers for Disease Control and Prevention, The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the funding agencies. Data Availability All data produced in the present study are available upon reasonable request to the authors Acknowledgements Nicaraguan samples: The SAGE study was made possible with the NIH grants R01AI127845 and: K24AI141744. Kenyan samples: We acknowledge the Duke Microbiome Core Facility for performing the DNA extractions, quality control, and preparing the 16S rRNA gene sequencing library; the Duke Sequencing and Genomic Technologies Shared Resource for sequencing the 16S rRNA gene library. We also thank the Duke Microbiome Core Facility for performing DNA BioAnalyzer quality assessments. Pakistan samples: SEEM study was a multi-national collaboration between Aga Khan University (Pakistan), Washington University in St. Louis, the University of Virginia, and Cincinnati Children’s Hospital Medical Center, funded by the Bill and Melinda Gates Foundation (2016-2019). Marty Meier provided technical assistance in stool DNA purification. Linda Adair, Jun Zeng, Brianna Petrone, Michelle Kirtley, provided valuable scientific guidance to this study. Finally, we thank all of the children who were participants in this study, as well as their parents. Without their time, effort, and participation, this study would not have occurred. Footnotes The author list of this manuscript was revised; Figure 4 and Figure S2 were revised; Supplemental figures 5 and 6, displaying alternate model results, were added; Acknowledgements section was expanded; The manuscript also underwent overall formatting and stylistic revisions, as well as corrections in areas where the text did not match the figures. Works Cited 1. ↵ Subramanian , S. , Huq , S. , Yatsunenko , T. , Haque , R. , Mahfuz , M. , Alam , M.A. , Benezra , A. , DeStefano , J. , Meier , M.F. , Muegge , B.D. , et al. ( 2014 ). Persistent gut microbiota immaturity in malnourished Bangladeshi children . Nature 510 , 417 – 421 . doi: 10.1038/nature13421 . OpenUrl CrossRef PubMed Web of Science 2. ↵ Bäckhed , F. , Roswall , J. , Peng , Y. , Feng , Q. , Jia , H. , Kovatcheva-Datchary , P. , Li , Y. , Xia , Y. , Xie , H. , Zhong , H. , et al. ( 2015 ). Dynamics and Stabilization of the Human Gut Microbiome during the First Year of Life . Cell Host Microbe 17 , 690 – 703 . doi: 10.1016/j.chom.2015.04.004 . OpenUrl CrossRef PubMed 3. ↵ Fahur Bottino , G. , Bonham , K.S. , Patel , F. , McCann , S. , Zieff , M. , Naspolini , N. , Ho , D. , Portlock , T. , Joos , R. , Midani , F.S. , et al. ( 2025 ). Early life microbial succession in the gut follows common patterns in humans across the globe . Nat. Commun . 16 , 660 . doi: 10.1038/s41467-025-56072-w . OpenUrl CrossRef PubMed 4. ↵ 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 . OpenUrl CrossRef PubMed 5. ↵ Bager , P. , Wohlfahrt , J. , and Westergaard , T . ( 2008 ). Caesarean delivery and risk of atopy and allergic disease: meta-analyses . Clin. Exp. Allergy J. Br. Soc. Allergy Clin. Immunol . 38 , 634 – 642 . doi: 10.1111/j.1365-2222.2008.02939.x . OpenUrl CrossRef PubMed Web of Science 6. ↵ Clausen , T.D. , Bergholt , T. , Bouaziz , O. , Arpi , M. , Eriksson , F. , Rasmussen , S. , Keiding , N. , and Løkkegaard , E.C . ( 2016 ). Broad-Spectrum Antibiotic Treatment and Subsequent Childhood Type 1 Diabetes: A Nationwide Danish Cohort Study . PloS One 11 , e0161654 . doi: 10.1371/journal.pone.0161654 . OpenUrl CrossRef PubMed 7. ↵ Bokulich , N.A. , Chung , J. , Battaglia , T. , Henderson , N. , Jay , M. , Li , H. , D. Lieber , A. , Wu , F. , Perez-Perez , G.I. , Chen , Y. , et al. ( 2016 ). Antibiotics, birth mode, and diet shape microbiome maturation during early life . Sci. Transl. Med . 8 , 343r a82 . doi: 10.1126/scitranslmed.aad7121 . OpenUrl CrossRef 8. ↵ Stewart , C.J. , Ajami , N.J. , O’Brien , J.L. , Hutchinson , D.S. , Smith , D.P. , Wong , M.C. , Ross , M.C. , Lloyd , R.E. , Doddapaneni , H. , Metcalf , G.A. , 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 . OpenUrl CrossRef PubMed 9. ↵ David , L.A. , Maurice , C.F. , Carmody , R.N. , Gootenberg , D.B. , Button , J.E. , Wolfe , B.E. , Ling , A.V. , Devlin , A.S. , Varma , Y. , Fischbach , M.A. , et al. ( 2014 ). Diet rapidly and reproducibly alters the human gut microbiome . Nature 505 , 559 – 563 . doi: 10.1038/nature12820 . OpenUrl CrossRef PubMed Web of Science 10. ↵ Krebs , N.F. , Sherlock , L.G. , Westcott , J. , Culbertson , D. , Hambidge , K.M. , Feazel , L.M. , Robertson , C.E. , and Frank , D.N . ( 2013 ). Effects of Different Complementary Feeding Regimens on Iron Status and Enteric Microbiota in Breastfed Infants . J. Pediatr . 163 , 416 – 423 .e4. doi: 10.1016/j.jpeds.2013.01.024 . OpenUrl CrossRef PubMed Web of Science 11. Qasem , W. , Azad , M.B. , Hossain , Z. , Azad , E. , Jorgensen , S. , Castillo San Juan , S. , Cai , C. , Khafipour , E. , Beta , T. , Roberts , L.J. , et al. ( 2017 ). Assessment of complementary feeding of Canadian infants: effects on microbiome & oxidative stress, a randomized controlled trial . BMC Pediatr . 17 , 54 . doi: 10.1186/s12887-017-0805-0 . OpenUrl CrossRef PubMed 12. Tang , M. , Frank , D. , Hendricks , A. , Ir , D. , and Krebs , N . ( 2019 ). Protein Intake During Early Complementary Feeding Affects the Gut Microbiota in U.S. Formula-fed Infants (FS04-03-19) . Nutr . 2019 Abstr. 3 , nzz048.FS04-03-19. doi: 10.1093/cdn/nzz048.FS04-03-19 . OpenUrl CrossRef 13. ↵ Plaza-Diaz , J. , Bernal , M.J. , Schutte , S. , Chenoll , E. , Genovés , S. , Codoñer , F.M. , Gil , A. , and Sanchez-Siles , L.M . ( 2021 ). Effects of Whole-Grain and Sugar Content in Infant Cereals on Gut Microbiota at Weaning: A Randomized Trial . Nutrients 13 , 1496 . doi: 10.3390/nu13051496 . OpenUrl CrossRef PubMed 14. ↵ Smith-Brown , P. , Morrison , M. , Krause , L. , and Davies , P.S.W . ( 2016 ). Dairy and plant based food intakes are associated with altered faecal microbiota in 2 to 3 year old Australian children . Sci. Rep . 6 , 32385 . doi: 10.1038/srep32385 . OpenUrl CrossRef PubMed 15. Smith-Brown , P. , Morrison , M. , Krause , L. , and Davies , P.S.W . ( 2019 ). Microbiota and Body Composition During the Period of Complementary Feeding . J. Pediatr. Gastroenterol. Nutr . 69 , 726 – 732 . doi: 10.1097/MPG.0000000000002490 . OpenUrl CrossRef PubMed 16. Thompson , A.L. , Monteagudo-Mera , A. , Cadenas , M.B. , Lampl , M.L. , and Azcarate-Peril , M.A . ( 2015 ). Milk- and solid-feeding practices and daycare attendance are associated with differences in bacterial diversity, predominant communities, and metabolic and immune function of the infant gut microbiome . Front. Cell. Infect. Microbiol . 5 . doi: 10.3389/fcimb.2015.00003 . OpenUrl CrossRef 17. ↵ Laursen Martin Frederik , Andersen Louise B. B. , Michaelsen Kim F. , Mølgaard Christian , Trolle Ellen , Bahl Martin Iain , Licht Tine Rask , and Suen Garret ( 2016 ). Infant Gut Microbiota Development Is Driven by Transition to Family Foods Independent of Maternal Obesity . mSphere 1 , e00069 – 15 . doi: 10.1128/mSphere.00069-15 . OpenUrl CrossRef 18. ↵ Olm , M.R. , Dahan , D. , Carter , M.M. , Merrill , B.D. , Yu , F.B. , Jain , S. , Meng , X. , Tripathi , S. , Wastyk , H. , Neff , N. , et al. ( 2022 ). Robust variation in infant gut microbiome assembly across a spectrum of lifestyles . Science 376 , 1220 – 1223 . doi: 10.1126/science.abj2972 . OpenUrl CrossRef PubMed 19. ↵ Kipnis , V. , Subar , A.F. , Midthune , D. , Freedman , L.S. , Ballard-Barbash , R. , Troiano , R.P. , Bingham , S. , Schoeller , D.A. , Schatzkin , A. , and Carroll , R.J . ( 2003 ). Structure of Dietary Measurement Error: Results of the OPEN Biomarker Study . Am. J. Epidemiol . 158 , 14 – 21 . doi: 10.1093/aje/kwg091 . OpenUrl CrossRef PubMed Web of Science 20. ↵ Subar , A.F. , Freedman , L.S. , Tooze , J.A. , Kirkpatrick , S.I. , Boushey , C. , Neuhouser , M.L. , Thompson , F.E. , Potischman , N. , Guenther , P.M. , Tarasuk , V. , et al. ( 2015 ). Addressing Current Criticism Regarding the Value of Self-Report Dietary Data12 . J. Nutr . 145 , 2639 – 2645 . doi: 10.3945/jn.115.219634 . OpenUrl Abstract / FREE Full Text 21. ↵ What We Eat in America, NHANES 2017-March 2020 Prepandemic, individuals 2 years and over (excluding breast-fed children) , day 1. 22. ↵ Burrows , T.L. , Martin , R.J. , and Collins , C.E . ( 2010 ). A Systematic Review of the Validity of Dietary Assessment Methods in Children when Compared with the Method of Doubly Labeled Water . J. Am. Diet. Assoc . 110 , 1501 – 1510 . doi: 10.1016/j.jada.2010.07.008 . OpenUrl CrossRef PubMed Web of Science 23. Davies , P.S.W. , Coward , W.A. , Gregory , J. , White , A. , and Mills , A . ( 1994 ). Total energy expenditure and energy intake in the pre-school child: a comparison . Br. J. Nutr . 72 , 13 – 20 . doi: 10.1079/BJN19940005 . OpenUrl CrossRef PubMed 24. Fisher , J.O. , Butte , N.F. , Mendoza , P.M. , Wilson , T.A. , Hodges , E.A. , Reidy , K.C. , and Deming , D . ( 2008 ). Overestimation of infant and toddler energy intake by 24-h recall compared with weighed food records . Am. J. Clin. Nutr . 88 , 407 – 415 . doi: 10.1093/ajcn/88.2.407 . OpenUrl Abstract / FREE Full Text 25. Lanigan , J.A. , Wells , J.C.K. , Lawson , M.S. , and Lucas , A . ( 2001 ). Validation of food diary method for assessment of dietary energy and macronutrient intake in infants and children aged 6–24 months . Eur. J. Clin. Nutr . 55 , 124 – 129 . doi: 10.1038/sj.ejcn.1601128 . OpenUrl CrossRef PubMed Web of Science 26. Ravelli , M.N. , and Schoeller , D.A . ( 2020 ). Traditional Self-Reported Dietary Instruments Are Prone to Inaccuracies and New Approaches Are Needed . Front. Nutr . 7 . doi: 10.3389/fnut.2020.00090 . OpenUrl CrossRef PubMed 27. ↵ Wallace , A. , Kirkpatrick , S.I. , Darlington , G. , and Haines , J . ( 2018 ). Accuracy of Parental Reporting of Preschoolers’ Dietary Intake Using an Online Self-Administered 24-h Recall . Nutrients 10 , 987 . doi: 10.3390/nu10080987 . OpenUrl CrossRef PubMed 28. ↵ Data FAO/WHO GIFT | Global Individual Food Consumption Data Tool GIFT . https://www.fao.org/gift-individual-food-consumption/data/en . 29. ↵ Global Diet Quality Project ( 2022 ). Measuring what the world eats: Insights from a new approach . (Geneva: Global Alliance for Improved Nutrition (GAIN ); Boston, MA : Harvard T.H. Chan School of Public Health, Department of Global Health and Population .). 30. ↵ Petrone , B.L. , Aqeel , A. , Jiang , S. , Durand , H.K. , Dallow , E.P. , McCann , J.R. , Dressman , H.K. , Hu , Z. , Tenekjian , C.B. , Yancy , W.S. , et al. ( 2023 ). Diversity of plant DNA in stool is linked to dietary quality, age, and household income . Proc. Natl. Acad. Sci . 120 , e2304441120 . doi: 10.1073/pnas.2304441120 . OpenUrl CrossRef PubMed 31. ↵ Loftus , T. ( 1995 ). Staple foods: what do people eat? In Dimensions of Need (Food and Agriculture organization of the United Nations) . 32. ↵ Mank , I. , Vandormael , A. , Traoré , I. , Ouédraogo , W.A. , Sauerborn , R. , and Danquah , I . ( 2020 ). Dietary habits associated with growth development of children aged < 5 years in the Nouna Health and Demographic Surveillance System, Burkina Faso . Nutr. J . 19 , 81 . doi: 10.1186/s12937-020-00591-3 . OpenUrl CrossRef PubMed 33. Steyn , N.P. , Nel , J.H. , Nantel , G. , Kennedy , G. , and Labadarios , D . ( 2006 ). Food variety and dietary diversity scores in children: are they good indicators of dietary adequacy? Public Health Nutr . 9 , 644 – 650 . doi: 10.1079/phn2005912 . OpenUrl CrossRef PubMed Web of Science 34. ↵ Mezgebo , K. , Lema , T.B. , and Neela , S . ( 2021 ). Food variety, dietary diversity scores and dietary quality of complementary foods consuming by 6-24 months aged children in Jimma town, southwest Ethiopia . Nutr. Food Sci . 51 , 323 – 344 . doi: 10.1108/NFS-02-2020-0033 . OpenUrl CrossRef 35. ↵ Dominguez-Bello , M.G. , Costello , E.K. , Contreras , M. , Magris , M. , Hidalgo , G. , Fierer , N. , and Knight , R . ( 2010 ). Delivery mode shapes the acquisition and structure of the initial microbiota across multiple body habitats in newborns . Proc. Natl. Acad. Sci . 107 , 11971 . doi: 10.1073/pnas.1002601107 . OpenUrl Abstract / FREE Full Text 36. ↵ Abdelsalam , N.A. , Hegazy , S.M. , and Aziz , R.K . The curious case of Prevotella copri . Gut Microbes 15 , 2249152 . doi: 10.1080/19490976.2023.2249152 . OpenUrl CrossRef PubMed 37. ↵ Johnson , K.V.-A. , and Burnet , P.W.J . ( 2016 ). Microbiome: Should we diversify from diversity? Gut Microbes 7 , 455 – 458 . doi: 10.1080/19490976.2016.1241933 . OpenUrl CrossRef PubMed 38. ↵ Starke , R. , Capek , P. , Morais , D. , Callister , S.J. , and Jehmlich , N . ( 2020 ). The total microbiome functions in bacteria and fungi . J. Proteomics 213 , 103623 . doi: 10.1016/j.jprot.2019.103623 . OpenUrl CrossRef 39. ↵ Huttenhower , C. , Gevers , D. , Knight , R. , Abubucker , S. , Badger , J.H. , Chinwalla , A.T. , Creasy , H.H. , Earl , A.M. , FitzGerald , M.G. , Fulton , R.S. , et al. ( 2012 ). Structure, function and diversity of the healthy human microbiome . Nature 486 , 207 – 214 . doi: 10.1038/nature11234 . OpenUrl CrossRef PubMed Web of Science 40. ↵ Branca , F. , Lartey , A. , Oenema , S. , Aguayo , V. , Stordalen , G.A. , Richardson , R. , Arvelo , M. , and Afshin , A . ( 2019 ). Transforming the food system to fight non-communicable diseases . BMJ 364 , l296 . doi: 10.1136/bmj.l296 . OpenUrl FREE Full Text 41. ↵ Arimond , M. , and Ruel , M.T . ( 2004 ). Dietary Diversity Is Associated with Child Nutritional Status: Evidence from 11 Demographic and Health Surveys . J. Nutr . 134 , 2579 – 2585 . doi: 10.1093/jn/134.10.2579 . OpenUrl Abstract / FREE Full Text 42. ↵ Kennedy , G.L. , Pedro , M.R. , Seghieri , C. , Nantel , G. , and Brouwer , I . ( 2007 ). Dietary diversity score is a useful indicator of micronutrient intake in non-breast-feeding Filipino children . J. Nutr . 137 , 472 – 477 . doi: 10.1093/jn/137.2.472 . OpenUrl Abstract / FREE Full Text 43. ↵ Roduit , C. , Frei , R. , Depner , M. , Schaub , B. , Loss , G. , Genuneit , J. , Pfefferle , P. , Hyvärinen , A. , Karvonen , A.M. , Riedler , J. , et al. ( 2014 ). Increased food diversity in the first year of life is inversely associated with allergic diseases . J. Allergy Clin. Immunol . 133 , 1056 – 1064 .e7. doi: 10.1016/j.jaci.2013.12.1044 . OpenUrl CrossRef Web of Science 44. ↵ Indicators for assessing infant and young child feeding practices: definitions and measurement methods ( 2021 ). 45. ↵ Lai , A. , Velez , I. , Ambikapathi , R. , Seng , K. , Levy , K. , Kowalsky , E. , Holcomb , D. , Konstantinidis , K.T. , Cumming , O. , and Brown , J . ( 2022 ). Independent and combined effects of nutrition and sanitation interventions on enteric pathogen carriage and child growth in rural Cambodia: a factorial cluster-randomised controlled trial . Preprint at medRxiv , doi: 10.1101/2021.05.21.21257546 10.1101/2021.05.21.21257546. OpenUrl Abstract / FREE Full Text 46. ↵ Song , R. , Click , E.S. , McCarthy , K.D. , Heilig , C.M. , Mchembere , W. , Smith , J.P. , Fajans , M. , Musau , S.K. , Okeyo , E. , Okumu , A. , et al. ( 2021 ). Sensitive and Feasible Specimen Collection and Testing Strategies for Diagnosing Tuberculosis in Young Children . JAMA Pediatr . 175 , e206069 . doi: 10.1001/jamapediatrics.2020.6069 . OpenUrl CrossRef 47. ↵ Becker-Dreps , S. , Allali , I. , Monteagudo , A. , Vilchez , S. , Hudgens , M.G. , Rogawski , E.T. , Carroll , I.M. , Zambrana , L.E. , Espinoza , F. , and Azcarate-Peril , M.A . ( 2015 ). Gut Microbiome Composition in Young Nicaraguan Children During Diarrhea Episodes and Recovery . Am. J. Trop. Med. Hyg . 93 , 1187 – 1193 . doi: 10.4269/ajtmh.15-0322 . OpenUrl Abstract / FREE Full Text 48. ↵ Study Details | Project HOPE 1000: Health Outcomes Pregnancy Exposures | ClinicalTrials.gov https://clinicaltrials.gov/study/NCT04417283 . 49. ↵ Martin , V.M. , Virkud , Y.V. , Dahan , E. , Seay , H.L. , Itzkovits , D. , Vlamakis , H. , Xavier , R. , Shreffler , W.G. , Yuan , Q. , and Yassour , M . ( 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 . OpenUrl CrossRef 50. ↵ Caporaso , J.G. , Lauber , C.L. , Walters , W.A. , Berg-Lyons , D. , Huntley , J. , Fierer , N. , Owens , S.M. , Betley , J. , Fraser , L. , Bauer , M. , et al. ( 2012 ). Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms . ISME J . 6 , 1621 – 1624 . doi: 10.1038/ismej.2012.8 . OpenUrl CrossRef PubMed Web of Science 51. ↵ Caporaso , J.G. , Lauber , C.L. , Walters , W.A. , Berg-Lyons , D. , Lozupone , C.A. , Turnbaugh , P.J. , Fierer , N. , and Knight , R . ( 2011 ). Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample . Proc. Natl. Acad. Sci . 108 , 4516 – 4522 . doi: 10.1073/pnas.1000080107 . OpenUrl Abstract / FREE Full Text 52. ↵ Maurice , C.F. , Haiser , H.J. , and Turnbaugh , P.J . ( 2013 ). Xenobiotics Shape the Physiology and Gene Expression of the Active Human Gut Microbiome . Cell 152 , 39 – 50 . doi: 10.1016/j.cell.2012.10.052 . OpenUrl CrossRef PubMed Web of Science 53. ↵ Petrone , B.L. , Aqeel , A. , Jiang , S. , Durand , H.K. , Dallow , E.P. , McCann , J.R. , Dressman , H.K. , Hu , Z. , Tenekjian , C.B. , Yancy , W.S. , et al. ( 2022 ). DNA metabarcoding captures dietary plant diversity in individuals and cohorts . Preprint at medRxiv , doi: 10.1101/2022.06.13.22276343 10.1101/2022.06.13.22276343. OpenUrl Abstract / FREE Full Text 54. ↵ Shehzad , W. , Riaz , T. , Nawaz , M.A. , Miquel , C. , Poillot , C. , Shah , S.A. , Pompanon , F. , Coissac , E. , and Taberlet , P . ( 2012 ). Carnivore diet analysis based on next-generation sequencing: application to the leopard cat (Prionailurus bengalensis) in Pakistan . Mol. Ecol . 21 , 1951 – 1965 . doi: 10.1111/j.1365-294X.2011.05424.x . OpenUrl CrossRef PubMed 55. ↵ De Barba , M. , Miquel , C. , Boyer , F. , Mercier , C. , Rioux , D. , Coissac , E. , and Taberlet , P. ( 2014 ). DNA metabarcoding multiplexing and validation of data accuracy for diet assessment: application to omnivorous diet . Mol. Ecol. Resour . 14 , 306 – 323 . doi: 10.1111/1755-0998.12188 . OpenUrl CrossRef PubMed 56. ↵ David , L.A. , Weil , A. , Ryan , E.T. , Calderwood , S.B. , Harris , J.B. , Chowdhury , F. , Begum , Y. , Qadri , F. , LaRocque , R.C. , and Turnbaugh , P.J . ( 2015 ). Gut Microbial Succession Follows Acute Secretory Diarrhea in Humans . mBio 6 , doi: 10.1128/mbio.00381-15 . 10.1128/mbio.00381-15. OpenUrl CrossRef 57. ↵ Schneider , J . ( 2021 ). Comprehensive coverage of human last meal components revealed by a forensic DNA metabarcoding approach | Scientific Reports. Sci. Rep . 11 . 58. Nations, U. Human Development Index (United Nations) . View the discussion thread. Back to top Previous Next Posted May 14, 2025. Download PDF Data/Code Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Dietary plant diversity predicts early life microbiome maturation Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Dietary plant diversity predicts early life microbiome maturation Teresa K. McDonald , Ammara Aqeel , Ben Neubert , Anna Bauer , Sharon Jiang , Olivia Osborne , Nolan Ives , Danting Jiang , Filemon Bucardo , Lester Gutiérrez , Luis Zambrana , Kirsten Jenkins , Jennifer Gilner , Javier Rodriguez , Amanda Lai , Jonathan P. Smith , Rinn Song , Kazi Ahsan , Sheraz Ahmed , Sanam Iram Soomro , Fayaz Umrani , Michael Barratt , Jeffrey I. Gordon , Asad Ali , Najeeha Iqbal , Jillian H. Hurst , Victoria Martin , Wayne Shreffler , Qian Yuan , Joe M. Brown , Neeraj K. Surana , Samuel Vilchez , Sylvia Becker-Dreps , Lawrence A David medRxiv 2025.02.28.25323117; doi: https://doi.org/10.1101/2025.02.28.25323117 Share This Article: Copy Citation Tools Dietary plant diversity predicts early life microbiome maturation Teresa K. McDonald , Ammara Aqeel , Ben Neubert , Anna Bauer , Sharon Jiang , Olivia Osborne , Nolan Ives , Danting Jiang , Filemon Bucardo , Lester Gutiérrez , Luis Zambrana , Kirsten Jenkins , Jennifer Gilner , Javier Rodriguez , Amanda Lai , Jonathan P. Smith , Rinn Song , Kazi Ahsan , Sheraz Ahmed , Sanam Iram Soomro , Fayaz Umrani , Michael Barratt , Jeffrey I. Gordon , Asad Ali , Najeeha Iqbal , Jillian H. Hurst , Victoria Martin , Wayne Shreffler , Qian Yuan , Joe M. Brown , Neeraj K. Surana , Samuel Vilchez , Sylvia Becker-Dreps , Lawrence A David medRxiv 2025.02.28.25323117; doi: https://doi.org/10.1101/2025.02.28.25323117 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Epidemiology Subject Areas All Articles Addiction Medicine (569) Allergy and Immunology (863) Anesthesia (300) Cardiovascular Medicine (4442) Dentistry and Oral Medicine (444) Dermatology (383) Emergency Medicine (609) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1511) Epidemiology (15230) Forensic Medicine (30) Gastroenterology (1126) Genetic and Genomic Medicine (6610) Geriatric Medicine (668) Health Economics (998) Health Informatics (4542) Health Policy (1370) Health Systems and Quality Improvement (1613) Hematology (543) HIV/AIDS (1266) Infectious Diseases (except HIV/AIDS) (15923) Intensive Care and Critical Care Medicine (1103) Medical Education (623) Medical Ethics (147) Nephrology (668) Neurology (6607) Nursing (346) Nutrition (999) Obstetrics and Gynecology (1146) Occupational and Environmental Health (957) Oncology (3338) Ophthalmology (974) Orthopedics (369) Otolaryngology (420) Pain Medicine (436) Palliative Medicine (130) Pathology (665) Pediatrics (1693) Pharmacology and Therapeutics (692) Primary Care Research (712) Psychiatry and Clinical Psychology (5448) Public and Global Health (9238) Radiology and Imaging (2202) Rehabilitation Medicine and Physical Therapy (1370) Respiratory Medicine (1196) Rheumatology (596) Sexual and Reproductive Health (714) Sports Medicine (530) Surgery (712) Toxicology (99) Transplantation (289) Urology (265) (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a01e0c55ae7ef047',t:'MTc3OTgxMTgwNw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();