Protocol for the Gut Bugs in Autism Trial: a double-blind randomised placebo-controlled trial of faecal microbiome transfer for the treatment of gastrointestinal symptoms in autistic adolescents and adults.

INTRODUCTION: Autism (formally autism spectrum disorder) encompasses a group of complex neurodevelopmental conditions, characterised by differences in communication and social interactions. Co-occurring chronic gastrointestinal symptoms are common among autistic individuals and can adversely affect their quality of life. This study aims to evaluate the efficacy of oral encapsulated faecal microbiome transfer (FMT) in improving gastrointestinal symptoms and well-being among autistic adolescents and adults. METHODS AND ANALYSIS: This double-blind, randomised, placebo-controlled trial will recruit 100 autistic adolescents and adults aged 16-45 years, who have mild to severe gastrointestinal symptoms (Gastrointestinal Symptoms Rating Scale (GSRS) score ≥2.0). We will also recruit eight healthy donors aged 18-32 years, who will undergo extensive clinical screening. Recipients will be randomised 1:1 to receive FMT or placebo, stratified by biological sex. Capsules will be administered over two consecutive days following an overnight bowel cleanse with follow-up assessments at 6, 12 and 26 weeks post-treatment. The primary outcome is GSRS score at 6 weeks. Other assessments include anthropometry, body composition, hair cortisol concentration, gut microbiome profile, urine/plasma gut-derived metabolites, plasma markers of gut inflammation/permeability and questionnaires on general well-being, sleep quality, physical activity, food diversity and treatment tolerability. Adverse events will be recorded and reviewed by an independent data monitoring committee. ETHICS AND DISSEMINATION: Ethics approval for the study was granted by the Central Health and Disability Ethics Committee on 24 August 2021 (reference number: 21/CEN/211). Results will be published in peer-reviewed journals and presented to both scientific and consumer group audiences. TRIAL REGISTRATION NUMBER: ACTRN12622000015741.

Horizontal gene transfer after faecal microbiota transplantation in adolescents with obesity.

BACKGROUND: Horizontal gene transfer (HGT) describes the transmission of DNA outside of direct ancestral lineages. The process is best characterised within the bacterial kingdom and can enable the acquisition of genetic traits that support bacterial adaptation to novel niches. The adaptation of bacteria to novel niches has particular relevance for faecal microbiota transplantation (FMT), a therapeutic procedure which aims to resolve gut-related health conditions of individuals, through transplanted gut microbiota from healthy donors. RESULTS: Three hundred eighty-one stool metagenomic samples from a placebo-controlled FMT trial for obese adolescents (the Gut Bugs Trial) were analysed for HGT, using two complementary methodologies. First, all putative HGT events, including historical HGT signatures, were quantified using the bioinformatics application WAAFLE. Second, metagenomic assembly and gene clustering were used to assess and quantify donor-specific genes transferred to recipients following the intervention. Both methodologies found no difference between the level of putative HGT events in the gut microbiomes of FMT and placebo recipients, post-intervention. HGT events facilitated by engrafted donor species in the FMT recipient gut at 6 weeks post-intervention were identified and characterised. Bacterial strains contributing to this subset of HGT events predominantly belonged to the phylum Bacteroidetes. Engraftment-dependent horizontally transferred genes were retained within recipient microbiomes at 12 and 26 weeks post-intervention. CONCLUSION: Our study suggests that novel microorganisms introduced into the recipient gut following FMT have no impact on the basal rate of HGT within the human gut microbiome. Analyses of further FMT studies are required to assess the generalisability of this conclusion across different FMT study designs and for the treatment of different gut-related conditions. Video Abstract.

An open-label pilot trial of faecal microbiome transfer to restore the gut microbiome in anorexia nervosa: protocol.

INTRODUCTION: Individuals with anorexia nervosa (AN) harbour distinct gut microbiomes compared with healthy individuals, which are sufficient to induce weight loss and anxiety-like behaviours when transplanted into germ-free mice. We hypothesise that faecal microbiome transfer (FMT) from healthy donors would help restore the gut microbiome of individuals with AN, which in turn, may aid patient recovery. METHODS: We aim to conduct an open-label pilot study in 20 females aged 16-32 years in Auckland, New Zealand who meet the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5) criteria for AN and have a body mass index 13-19 kg/m2. We will recruit four healthy, lean, female donors, aged 18-32 years, who will undergo extensive clinical screening prior to stool donation. Faecal microbiota will be harvested from donors and double encapsulated in delayed release, acid-resistant capsules. All participants will receive a single course of 20 FMT capsules (five from each donor) which they can choose to take over two or four consecutive days. Stool and blood samples will be collected from participants over a period of 3 months to assess their gut microbiome profile, metabolome, levels of intestinal inflammation and nutritional status. Our primary outcome is a shift in the gut microbiome composition at 3 weeks post-FMT (Bray-Curtis dissimilarity). We will also monitor participants' body composition (whole-body dual-energy X-ray absorptiometry scans), eating disorder psychopathology, mental health and assess their views on, and tolerability of, treatment. All adverse events will be recorded and reviewed by an independent data monitoring committee. ETHICS AND DISSEMINATION: Ethics approval was provided by the Central Health and Disability Ethics Committee (Ministry of Health, New Zealand, 21/CEN/212). Results will be published in peer-reviewed journals and presented to both scientific and consumer group audiences. TRIAL REGISTRATION NUMBER: ACTRN12621001504808.

Centenarians have a diverse gut virome with the potential to modulate metabolism and promote healthy lifespan.

Distinct gut microbiome ecology may be implicated in the prevention of aging-related diseases as it influences systemic immune function and resistance to infections. Yet, the viral component of the microbiome throughout different stages in life remains unexplored. Here we present a characterization of the centenarian gut virome using previously published metagenomes from 195 individuals from Japan and Sardinia. Compared with gut viromes of younger adults (>18 yr) and older individuals (>60 yr), centenarians had a more diverse virome including previously undescribed viral genera, such as viruses associated with Clostridia. A population shift towards higher lytic activity was also observed. Finally, we investigated phage-encoded auxiliary functions that influence bacterial physiology, which revealed an enrichment of genes supporting key steps in sulfate metabolic pathways. Phage and bacterial members of the centenarian microbiome displayed an increased potential for converting methionine to homocysteine, sulfate to sulfide and taurine to sulfide. A greater metabolic output of microbial hydrogen sulfide in centenarians may in turn support mucosal integrity and resistance to pathobionts.

Impact of exposure to per- and polyfluoroalkyl substances on fecal microbiota composition in mother-infant dyads.

There is growing evidence suggesting that chemical exposure alters gut microbiota composition. However, not much is known about the impact of per- and polyfluoroalkyl substances (PFAS) on the gut microbial community. Here, in a mother-infant study, we set out to identify the gut bacterial species that associate with chemical exposure before (maternal) and after (maternal, infant) birth. Paired serum and stool samples were collected from mother-infant dyads (n = 30) in a longitudinal setting. PFAS were quantified in maternal serum to examine their associations with the microbial compositions (determined by shotgun metagenomic sequencing) in mothers and infants. High maternal exposure to PFAS was consistently associated with increased abundance of Methanobrevibacter smithii in maternal stool. Among individual PFAS compounds, PFOS and PFHpS showed the strongest association with M. smithii. However, maternal total PFAS exposure associated only weakly with the infant microbiome. Our findings suggest that PFAS exposure affects the composition of the adult gut microbiome.

Comprehensive Functional Annotation of Metagenomes and Microbial Genomes Using a Deep Learning-Based Method.

Comprehensive protein function annotation is essential for understanding microbiome-related disease mechanisms in the host organisms. However, a large portion of human gut microbial proteins lack functional annotation. Here, we have developed a new metagenome analysis workflow integrating de novo genome reconstruction, taxonomic profiling, and deep learning-based functional annotations from DeepFRI. This is the first approach to apply deep learning-based functional annotations in metagenomics. We validate DeepFRI functional annotations by comparing them to orthology-based annotations from eggNOG on a set of 1,070 infant metagenomes from the DIABIMMUNE cohort. Using this workflow, we generated a sequence catalogue of 1.9 million nonredundant microbial genes. The functional annotations revealed 70% concordance between Gene Ontology annotations predicted by DeepFRI and eggNOG. DeepFRI improved the annotation coverage, with 99% of the gene catalogue obtaining Gene Ontology molecular function annotations, although they are less specific than those from eggNOG. Additionally, we constructed pangenomes in a reference-free manner using high-quality metagenome-assembled genomes (MAGs) and analyzed the associated annotations. eggNOG annotated more genes on well-studied organisms, such as Escherichia coli, while DeepFRI was less sensitive to taxa. Further, we show that DeepFRI provides additional annotations in comparison to the previous DIABIMMUNE studies. This workflow will contribute to novel understanding of the functional signature of the human gut microbiome in health and disease as well as guiding future metagenomics studies. IMPORTANCE The past decade has seen advancement in high-throughput sequencing technologies resulting in rapid accumulation of genomic data from microbial communities. While this growth in sequence data and gene discovery is impressive, the majority of microbial gene functions remain uncharacterized. The coverage of functional information coming from either experimental sources or inferences is low. To solve these challenges, we have developed a new workflow to computationally assemble microbial genomes and annotate the genes using a deep learning-based model DeepFRI. This improved microbial gene annotation coverage to 1.9 million metagenome-assembled genes, representing 99% of the assembled genes, which is a significant improvement compared to 12% Gene Ontology term annotation coverage by commonly used orthology-based approaches. Importantly, the workflow supports pangenome reconstruction in a reference-free manner, allowing us to analyze the functional potential of individual bacterial species. We therefore propose this alternative approach combining deep-learning functional predictions with the commonly used orthology-based annotations as one that could help us uncover novel functions observed in metagenomic microbiome studies.

Sequence-structure-function relationships in the microbial protein universe.

For the past half-century, structural biologists relied on the notion that similar protein sequences give rise to similar structures and functions. While this assumption has driven research to explore certain parts of the protein universe, it disregards spaces that don't rely on this assumption. Here we explore areas of the protein universe where similar protein functions can be achieved by different sequences and different structures. We predict ~200,000 structures for diverse protein sequences from 1,003 representative genomes across the microbial tree of life and annotate them functionally on a per-residue basis. Structure prediction is accomplished using the World Community Grid, a large-scale citizen science initiative. The resulting database of structural models is complementary to the AlphaFold database, with regards to domains of life as well as sequence diversity and sequence length. We identify 148 novel folds and describe examples where we map specific functions to structural motifs. We also show that the structural space is continuous and largely saturated, highlighting the need for a shift in focus across all branches of biology, from obtaining structures to putting them into context and from sequence-based to sequence-structure-function based meta-omics analyses.

Addressing antibiotic resistance: computational answers to a biological problem?

The increasing prevalence of infections caused by antibiotic-resistant bacteria is a global healthcare crisis. Understanding the spread of resistance is predicated on the surveillance of antibiotic resistance genes within an environment. Bioinformatics and artificial intelligence (AI) methods applied to metagenomic sequencing data offer the capacity to detect known and infer yet-unknown resistance mechanisms, and predict future outbreaks of antibiotic-resistant infections. Machine learning methods, in particular, could revive the waning antibiotic discovery pipeline by helping to predict the molecular structure and function of antibiotic resistance compounds, and optimising their interactions with target proteins. Consequently, AI has the capacity to play a central role in guiding antibiotic stewardship and future clinical decision-making around antibiotic resistance.

The fecal microbiotas of women of Pacific and New Zealand European ethnicities are characterized by distinctive enterotypes that reflect dietary intakes and fecal water content.

Obesity is a complex, multifactorial condition that is an important risk factor for noncommunicable diseases including cardiovascular disease and type 2 diabetes. While prevention and management require a healthy and energy balanced diet and adequate physical activity, the taxonomic composition and functional attributes of the colonic microbiota may have a supplementary role in the development of obesity. The taxonomic composition and metabolic capacity of the fecal microbiota of 286 women, resident in Auckland New Zealand, was determined by metagenomic analysis. Associations with BMI (obese, nonobese), body fat composition, and ethnicity (Pacific, n = 125; NZ European women [NZE], n = 161) were assessed using regression analyses. The fecal microbiotas were characterized by the presence of three distinctive enterotypes, with enterotype 1 represented in both Pacific and NZE women (39 and 61%, respectively), enterotype 2 mainly in Pacific women (84 and 16%) and enterotype 3 mainly in NZE women (13 and 87%). Enterotype 1 was characterized mainly by the relative abundances of butyrate producing species, Eubacterium rectale and Faecalibacterium prausnitzii, enterotype 2 by the relative abundances of lactic acid producing species, Bifidobacterium adolescentis, Bifidobacterium bifidum, and Lactobacillus ruminis, and enterotype 3 by the relative abundances of Subdoligranulum sp., Akkermansia muciniphila, Ruminococcus bromii, and Methanobrevibacter smithii. Enterotypes were also associated with BMI, visceral fat %, and blood cholesterol. Habitual food group intake was estimated using a 5 day nonconsecutive estimated food record and a 30 day, 220 item semi-quantitative Food Frequency Questionnaire. Higher intake of 'egg' and 'dairy' products was associated with enterotype 3, whereas 'non-starchy vegetables', 'nuts and seeds' and 'plant-based fats' were positively associated with enterotype 1. In contrast, these same food groups were inversely associated with enterotype 2. Fecal water content, as a proxy for stool consistency/colonic transit time, was associated with microbiota taxonomic composition and gene pools reflective of particular bacterial biochemical pathways. The fecal microbiotas of women of Pacific and New Zealand European ethnicities are characterized by distinctive enterotypes, most likely due to differential dietary intake and fecal consistency/colonic transit time. These parameters need to be considered in future analyses of human fecal microbiotas.

Finnish children carrying the high-risk HLA genotype have a 45-fold increased risk of type 1 diabetes compared to peers with neutral or protective genotypes.

The association between HLA genotypes and type 1 diabetes is well known. We set out to examine incidence rates and ratios of type 1 diabetes depending on the risk afflicted by HLA genotype. Children with the high-risk genotype have a 45-fold disease risk compared to peers with neutral or protective genotypes.

Mobile genetic elements from the maternal microbiome shape infant gut microbial assembly and metabolism.

The perinatal period represents a critical window for cognitive and immune system development, promoted by maternal and infant gut microbiomes and their metabolites. Here, we tracked the co-development of microbiomes and metabolomes from late pregnancy to 1 year of age using longitudinal multi-omics data from a cohort of 70 mother-infant dyads. We discovered large-scale mother-to-infant interspecies transfer of mobile genetic elements, frequently involving genes associated with diet-related adaptations. Infant gut metabolomes were less diverse than maternal but featured hundreds of unique metabolites and microbe-metabolite associations not detected in mothers. Metabolomes and serum cytokine signatures of infants who received regular-but not extensively hydrolyzed-formula were distinct from those of exclusively breastfed infants. Taken together, our integrative analysis expands the concept of vertical transmission of the gut microbiome and provides original insights into the development of maternal and infant microbiomes and metabolomes during late pregnancy and early life.

A distinct clade of Bifidobacterium longum in the gut of Bangladeshi children thrives during weaning.

The gut microbiome has an important role in infant health and development. We characterized the fecal microbiome and metabolome of 222 young children in Dhaka, Bangladesh during the first two years of life. A distinct Bifidobacterium longum clade expanded with introduction of solid foods and harbored enzymes for utilizing both breast milk and solid food substrates. The clade was highly prevalent in Bangladesh, present globally (at lower prevalence), and correlated with many other gut taxa and metabolites, indicating an important role in gut ecology. We also found that the B. longum clades and associated metabolites were implicated in childhood diarrhea and early growth, including positive associations between growth measures and B. longum subsp. infantis, indolelactate and N-acetylglutamate. Our data demonstrate geographic, cultural, seasonal, and ecological heterogeneity that should be accounted for when identifying microbiome factors implicated in and potentially benefiting infant development.

Dysregulation of secondary bile acid metabolism precedes islet autoimmunity and type 1 diabetes.

The gut microbiota is crucial in the regulation of bile acid (BA) metabolism. However, not much is known about the regulation of BAs during progression to type 1 diabetes (T1D). Here, we analyzed serum and stool BAs in longitudinal samples collected at 3, 6, 12, 18, 24, and 36 months of age from children who developed a single islet autoantibody (AAb) (P1Ab; n = 23) or multiple islet AAbs (P2Ab; n = 13) and controls (CTRs; n = 38) who remained AAb negative. We also analyzed the stool microbiome in a subgroup of these children. Factor analysis showed that age had the strongest impact on both BA and microbiome profiles. We found that at an early age, systemic BAs and microbial secondary BA pathways were altered in the P2Ab group compared with the P1Ab and CTR groups. Our findings thus suggest that dysregulated BA metabolism in early life may contribute to the risk and pathogenesis of T1D.

Strain-specific impacts of probiotics are a significant driver of gut microbiome development in very preterm infants.

The development of the gut microbiome from birth plays important roles in short- and long-term health, but factors influencing preterm gut microbiome development are poorly understood. In the present study, we use metagenomic sequencing to analyse 1,431 longitudinal stool samples from 123 very preterm infants (<32 weeks' gestation) who did not develop intestinal disease or sepsis over a study period of 10 years. During the study period, one cohort had no probiotic exposure whereas two cohorts were given different probiotic products: Infloran (Bifidobacterium bifidum and Lactobacillus acidophilus) or Labinic (B. bifidum, B. longum subsp. infantis and L. acidophilus). Mothers' own milk, breast milk fortifier, antibiotics and probiotics were significantly associated with the gut microbiome, with probiotics being the most significant factor. Probiotics drove microbiome transition into different preterm gut community types (PGCTs), each enriched in a different Bifidobacterium sp. and significantly associated with increased postnatal age. Functional analyses identified stool metabolites associated with PGCTs and, in preterm-derived organoids, sterile faecal supernatants impacted intestinal, organoid monolayer, gene expression in a PGCT-specific manner. The present study identifies specific influencers of gut microbiome development in very preterm infants, some of which overlap with those impacting term infants. The results highlight the importance of strain-specific differences in probiotic products and their impact on host interactions in the preterm gut.

A gut microbial peptide and molecular mimicry in the pathogenesis of type 1 diabetes.

Type 1 diabetes (T1D) is an autoimmune disease characterized by the destruction of pancreatic ß-cells. One of the earliest aspects of this process is the development of autoantibodies and T cells directed at an epitope in the B-chain of insulin (insB:9-23). Analysis of microbial protein sequences with homology to the insB:9-23 sequence revealed 17 peptides showing >50% identity to insB:9-23. Of these 17 peptides, the hprt4-18 peptide, found in the normal human gut commensal Parabacteroides distasonis, activated both human T cell clones from T1D patients and T cell hybridomas from nonobese diabetic (NOD) mice specific to insB:9-23. Immunization of NOD mice with P. distasonis insB:9-23 peptide mimic or insB:9-23 peptide verified immune cross-reactivity. Colonization of female NOD mice with P. distasonis accelerated the development of T1D, increasing macrophages, dendritic cells, and destructive CD8+ T cells, while decreasing FoxP3+ regulatory T cells. Western blot analysis identified P. distasonis-reacting antibodies in sera of NOD mice colonized with P. distasonis and human T1D patients. Furthermore, adoptive transfer of splenocytes from P. distasonis-treated mice to NOD/SCID mice enhanced disease phenotype in the recipients. Finally, analysis of human children gut microbiome data from a longitudinal DIABIMMUNE study revealed that seroconversion rates (i.e., the proportion of individuals developing two or more autoantibodies) were consistently higher in children whose microbiome harbored sequences capable of producing the hprt4-18 peptide compared to individuals who did not harbor it. Taken together, these data demonstrate the potential role of a gut microbiota-derived insB:9-23-mimic peptide as a molecular trigger of T1D pathogenesis.

Transcription shifts in gut bacteria shared between mothers and their infants.

The infant gut microbiome contains a portion of bacteria that originate from the maternal gut. In the infant gut these bacteria encounter a new metabolic environment that differs from the adult gut, consequently requiring adjustments in their activities. We used pilot community RNA sequencing data (metatranscriptomes) from ten mother-infant dyads participating in the NiPPeR Study to characterize bacterial gene expression shifts following mother-to-infant transmission. Maternally-derived bacterial strains exhibited large scale gene expression shifts following the transmission to the infant gut, with 12,564 activated and 14,844 deactivated gene families. The implicated genes were most numerous and the magnitude shifts greatest in Bacteroides spp. This pilot study demonstrates environment-dependent, strain-specific shifts in gut bacteria function and underscores the importance of metatranscriptomic analysis in microbiome studies.

Higher circulating EGF levels associate with a decreased risk of IgE sensitization in young children.

BACKGROUND: Decreased exposure to microbial agents in industrialized countries and urban living areas is considered as a risk factor of developing immune-mediated diseases, such as allergies and asthma. Epithelial surfaces in the gastrointestinal and respiratory tracts and in the skin constitute the primary areas in contact with the environmental microbial load. METHODS: We analyzed the levels of 30 cytokines and growth factors in serum or plasma as markers of the immune maturation in the participants in the DIABIMMUNE study from Russian Karelia (n = 60), Estonia (n = 83) and Finland (n = 89), three neighboring countries with remarkable differences in the incidences of allergies, asthma and autoimmune diseases. RESULTS: We observed an upregulation of T helper cell signature cytokines during the first 12 months of life, reflecting natural development of adaptive immune responses. During the first years of life, circulating concentrations of epidermal growth factor (EGF) were significantly higher, especially in Russian children compared with Finnish children. The children who developed IgE sensitization showed lower levels of EGF than those without such responses. CONCLUSION: Our results suggest that low circulating EGF levels associate with the risk of allergies possibly via the effects on the epithelial integrity and mucosal homeostasis.

Effect of prophylactic dextrose gel on the neonatal gut microbiome.

OBJECTIVE: To determine the effect of prophylactic dextrose gel on the infant gut microbiome. DESIGN: Observational cohort study nested in a randomised trial. SETTING: Three maternity hospitals in New Zealand. PATIENTS: Infants at risk of neonatal hypoglycaemia whose parents consented to participation in the hypoglycaemia Prevention in newborns with Oral Dextrose trial (hPOD). Infants were randomised to receive prophylactic dextrose gel or placebo gel, or were not randomised and received no gel (controls). Stool samples were collected on days 1, 7 and 28. MAIN OUTCOME MEASURES: The primary outcome was microbiome beta-diversity at 4 weeks. Secondary outcomes were beta-diversity, alpha-diversity, bacterial DNA concentration, microbial community stability and relative abundance of individual bacterial taxa at each time point. RESULTS: We analysed 434 stool samples from 165 infants using 16S rRNA gene amplicon sequencing. There were no differences between groups in beta-diversity at 4 weeks (p=0.49). There were also no differences between groups in any other microbiome measures including beta-diversity (p=0.53 at day 7), alpha-diversity (p=0.46 for day 7 and week 4), bacterial DNA concentration (p=0.91), microbial community stability (p=0.52) and microbial relative abundance at genus level. There was no evidence that exposure to any dextrose gel (prophylaxis or treatment) had any effect on the microbiome. Mode of birth, type of milk fed, hospital of birth and ethnicity were all associated with differences in the neonatal microbiome. CONCLUSIONS: Clinicians and consumers can be reassured that dextrose gel used for prophylaxis or treatment of neonatal hypoglycaemia does not alter the neonatal gut microbiome. TRIAL REGISTRATION NUMBER: 12614001263684.

Oral administration of maternal vaginal microbes at birth to restore gut microbiome development in infants born by caesarean section: A pilot randomised placebo-controlled trial.

BACKGROUND: Birth by caesarean section (CS) is associated with aberrant gut microbiome development and greater disease susceptibility later in life. We investigated whether oral administration of maternal vaginal microbiota to infants born by CS could restore their gut microbiome development in a pilot single-blinded, randomised placebo-controlled trial (Australian New Zealand Clinical Trials Registry, ACTRN12618000339257). METHODS: Pregnant women scheduled for a CS underwent comprehensive antenatal pathogen screening. At birth, healthy neonates were randomised to receive a 3 ml solution of either maternal vaginal microbes (CS-seeded, n = 12) or sterile water (CS-placebo, n = 13). Vaginally-born neonates were used as the reference control (VB, n = 22). Clinical assessments occurred within the first 2 h of birth, and at 1 month and 3 months of age. Infant stool samples and maternal vaginal extracts from CS women underwent shotgun metagenomic sequencing. The primary outcome was gut microbiome composition at 1 month of age. Secondary outcomes included maternal strain engraftment, functional potential of the gut microbiome, anthropometry, body composition, and adverse events. FINDINGS: Despite the presence of viable microbial cells within transplant solutions, there were no observed differences in gut microbiome composition or functional potential between CS-seeded and CS-placebo infants at 1 month or 3 months of age. Both CS groups displayed the characteristic signature of low Bacteroides abundance, which contributed to a number of biosynthesis pathways being underrepresented when compared with VB microbiomes. Maternal vaginal strain engraftment was rare. Vaginal seeding had no observed effects on anthropometry or body composition. There were no serious adverse events associated with treatment. INTERPRETATION: Our pilot findings question the value of vaginal seeding given that oral administration of maternal vaginal microbiota did not alter early gut microbiome development in CS-born infants. The limited colonisation of maternal vaginal strains suggest that other maternal sources, such as the perianal area, may play a larger role in seeding the neonatal gut microbiome. FUNDING: Health Research Council of New Zealand, A Better Start - National Science Challenge.

Structure-based protein function prediction using graph convolutional networks.

The rapid increase in the number of proteins in sequence databases and the diversity of their functions challenge computational approaches for automated function prediction. Here, we introduce DeepFRI, a Graph Convolutional Network for predicting protein functions by leveraging sequence features extracted from a protein language model and protein structures. It outperforms current leading methods and sequence-based Convolutional Neural Networks and scales to the size of current sequence repositories. Augmenting the training set of experimental structures with homology models allows us to significantly expand the number of predictable functions. DeepFRI has significant de-noising capability, with only a minor drop in performance when experimental structures are replaced by protein models. Class activation mapping allows function predictions at an unprecedented resolution, allowing site-specific annotations at the residue-level in an automated manner. We show the utility and high performance of our method by annotating structures from the PDB and SWISS-MODEL, making several new confident function predictions. DeepFRI is available as a webserver at https://beta.deepfri.flatironinstitute.org/ .