New Insights into the Composition of the Early-Life Gut Microbiome: A Metagenomic Analysis of Fecal Samples in Children Up to 18 Months of Age Authors: Egija Zelca, Dita Gudra, Daiga Karklina, Maija Ustinova, Megija Lunge, Davids Fridmanis, Ingrida Rumba-Rozenfelde, Ilva Daugule Published: 2026 ABSTRACT PurposeTo determine the gut microbiome composition and associated factors in infants.MethodsCross-sectional study was conducted in primary healthcare centers including healthy children up to 18 months of age. Parents of children answered a questionnaire and bring a fecal sample of their child. Fecal samples were analyzed using shotgun sequencing. The relative abundances of taxonomic profile were detected and compared with associated factors.ResultsIn the study were included 91 children. The most abundant taxonomic units belonged to phyla: Bacillota, Bacteroidota, Actinobacteria. In contrast, the most abundant species were Bifidobacterium longum and Bifidobacterium breve, demonstrating an age-appropriate microbial composition. The viral fraction mainly consisted of bacteriophages, with the most abundant viral family being Tospoviridae. Nevertheless, more than 50 rather new taxonomic entities were observed to cause shifts in the taxonomic units of the infant gut microbiome in association with perinatal antibacterial therapy, type of delivery, feeding type in the first 6 months of life. Higher relative abundance of Kitasatospora spp. MMS16 BH015 was associated with C-section and antibacterial treatment during delivery. A total of 302 antimicrobial resistance genes were identified, the most frequently encountered genes were associated with glycopeptide antibiotic resistance: vanG, vanT, vanW, vanY and adeF. The distribution of ARGs did not differ among different age groups.ConclusionMetagenomic sequencing provides broader insights into the composition of the gut microbiome, emphasizing the presence of unusual species associated with environmental and perinatal factors. Perinatal antibacterial treatment may have a greater impact on the infant microbiome than was previously estimated. INTRODUCTION With the development of culture-independent approaches, such as metagenomics, which reveal the presence of microorganisms based on their DNA sequences, detailed analyses of the gut microbiome composition and diversity have recently become possible [1]. Microorganisms in the human gut may further influence life and could be crucial for proper development of the immune system during the neonatal period [2]. The gut microbiome of infants is highly dynamic during the first year of life. The initial composition of the intestinal microbiome of an infant is influenced by many mother-infant and environmental factors such as the type of delivery, feeding pattern, gestational age, and use of antibiotics and prebiotics [3]. Although there is a high degree of intra-individual variability in the microbial composition of infants, the colonization of the gastrointestinal tract is typically initiated by the presence of facultative anaerobes, such as Enterobacteriaceae and Bifidobacterium, and with some lactic acid bacteria. After the introduction of solid food, bacterial population density increases, resulting in a shift in the environment that favors the growth of strict anaerobes, including Bifidobacterium, Firmicutes, and Bacteroides [45]. This change in the environment and competition between microbial species leads to a decrease in the number of facultative anaerobes. From this point onwards, the microbiome composition approaches an adult-like state. However, there is substantial variability among individuals in the first year of life, and an adult-like state is usually attained in the second and third years of life [5]. In addition to bacteria, a substantial number of viruses co-colonize the human gut and are collectively referred to as gut viromes. Furthermore, intestinal phages can influence the gut bacteriome function by regulating bacterial metabolism and viability, as well as by regulating the innate and adaptive immunity of the host. Therefore, evidence is accumulating to highlight the protective immunomodulatory activity of gut phages, which may mediate the beneficial effects of fecal microbiota transplantation. Similarly, the abundances of viral and phage communities in the gut change during infancy and childhood. In the first months of an infant’s life, the large number and variety of bacteriophages mainly consist of viruses belonging to the order Caudovirales. The richness of phages decreases over the first 2 years of life, and viruses belonging to the family Microviridae dominate. Nevertheless, the factors driving the shifts in bacteriophage communities during early life remain unclear [67]. However, because of their ability to alter the abundance of specific bacterial taxa to the point where they have been suggested as antimicrobial agents and therapeutic alternatives to antibiotics [8], it is crucial to understand the factors that influence phage communities during childhood. Antimicrobial resistance is a global health problem. In the early stages of life, antibiotic resistance can result from the horizontal and/or vertical transfer of antimicrobial resistance genes (ARGs) as well as from antibiotic exposure in infants [9]. The composition of the gut microbiome determines the presence of ARGs because most of these genes are carried by a few specific taxa. The most abundant ARGs in the infant gut microbiota belong to Escherichia, Klebsiella, Citrobacter, and Enterobacter species, whereas Bifidobacterium spp. might have a beneficial effect in reducing the ARG load in the infant gut. Alterations in the composition of the gut microbiome in early life could reportedly be a risk factor for metabolic and immune diseases [1011]. The association between specific entities at the species level, viral components, and ARGs, with several factors affecting the development of the gut microbiome, as well as the composition of the core microbiome, has not yet been elucidated. Therefore, we aimed to determine the composition of the gut microbiome and its influencing factors in Latvian infants using high-throughput metagenomic sequencing. Additionally, we determined the prevalence of ARGs in the infant microbiome. MATERIALS AND METHODS Study design A cross-sectional analysis was performed within the framework of a larger study “Gut microbiome composition in relation to extrinsic factors and health status in infants and toddlers,” conducted in primary healthcare centers in Latvia during 2020–2023. Parents of healthy children were asked to complete a questionnaire and collect fecal samples. DNA was isolated from fecal samples, and shotgun metagenomic sequencing analyses were performed. The relative abundances of taxonomic entities were determined and compared between groups of children with different environmental and medical conditions. Participant recruitment Healthy infants and toddlers aged up to 18 months were invited to participate in the study during their visits to a doctor for regular checkups and/or vaccinations. After obtaining signed informed consent from the child’s parent/caregiver, he/she was asked to answer a questionnaire and bring a fecal sample from their children. This study excluded infants and toddlers with acute illness or those undergoing antibacterial therapy at the time of sampling. The participants were divided into two age groups: the age group 0–6 months consisted of 38 infants, while the group 7–18 months comprised of 53 children. Questionnaire The parents completed a structured questionnaire during face-to-face interviews, which included questions about demographic data, perinatal anamnesis, infant feeding patterns, environmental and socioeconomic factors, history of antibacterial therapy, previous or present diseases, and family anamnesis (Supplementary Table 1). Sample collection Fecal samples were collected by the parents using an OMNIgene gut microbiome DNA collection kit that provided an opportunity to collect and stabilize DNA for gut microbiome profile analyses [12]. Parents were instructed about usage of the collection device by the researchers, but to ensure that the collection procedure was performed according to the instructions, it was also provided in written form. Parents collected fecal samples in circumstances that were comfortable for both them and the child. Following collection, samples were stored at room temperature (according to manufacturer’s instructions, fecal samples within OMNIgene container can be stored at room temperature for up to 60 days) and after their delivery to the healthcare office, they were transferred to a −80°C freezer. The collected containers were transported on dry ice to the research laboratory at the Latvian Biomedical Research and Study Centre and stored at −80°C until further processing. DNA extraction, library preparation and sequencing DNA was isolated from fecal samples using the FastDNA SPIN Kit for Soil according to the manufacturer’s guidelines. DNA samples for shotgun metagenomic analysis were normalized to an initial library input of 500 ng and sheared using a Covaris S220 focused ultrasonicator to reach an average fragment size of 400 bp. Libraries were further prepared using MGIEasy Universal DNA Library Prep Set V1.0 (MGI Tech Co.), according to the manufacturer’s recommendations. Quality control of the libraries was assessed using the Qubit High Sensitivity dsDNA assay kit on a Qubit 2.0 instrument and the Agilent High Sensitivity DNA kit on an Agilent 2100 Bioanalyzer. The sequencing depth was calculated to achieve approximately 20 million reads per sample (paired-end, read length 150 bp). Libraries were sequenced using a DNBSEQ-G400 sequencer and DNBSEQ-G400RS High-Throughput Sequencing Set PE150 (MGI Tech Co.), according to the standard workflow. Sequence data analysis Quality control and trimming of the obtained paired-end reads were performed using FastQC and Trimmomatic v0.39 [13], with a quality threshold of 30 and a minimum read length of 36. Quality-filtered sequences were then aligned to the human genome reference GRCh38 (hg38, UID:883148), and sequences matching the human genome were removed using Bowtie2 v.2.3.5.1. The taxonomic profile of the metagenomic dataset was assigned using MetaCache [14] and RefSeq database release 202 [15]. De novo read assembly into scaffolds was performed using the MetaSPAdes [16] assembler. The generated assembly was evaluated using metaQuast software [17]. Scaffolds were used to predict the resistome profile in the samples using The Resistance Gene Identifier (RGI) v.5.1.1, Comprehensive Antibiotic Resistance Database (CARD) and DIAMOND [18] alignment tool. The results were gathered for each sample using the heatmap function of the RGI by organizing resistance genes based on the resistance mechanism, gene family, and drug class. Hierarchical clustering was performed to cluster the samples based on their similarities. Statistical analyses Metadata and metagenomic data were analyzed using both MedCalc® and RStudio. Only sequence reads of 200 bp or more were used for data analysis. The mean or median relative abundances with 95% confidence intervals (CI) of the taxonomic entries provided by MetaCache were visualized using the microViz v.0.10.7 package. Alpha diversity measurement, the Shannon index, was calculated using the vegan v.2.6-2 and visualized using the phyloseq v.1.42.0 package within the R environment. To assess the significance of alpha diversity measurements between the patient groups, the Wilcoxon rank-sum test and Holm p-value adjustment method were performed using the vegan package. A p-value<0.05 was considered significant. Relative abundances among the entities were compared among groups of infants with the presence of different factors (previous maternal antibacterial treatment, type of delivery, breastfeeding and formula feeding, infants who had or had not received antibacterial therapy, living place, previous diseases, and antibacterial therapy), using the Wilcoxon rank sum test (MedCalc software) and multiple regression analyses; p-value<0.005 was considered significant. The prevalence of ARGs was analyzed using descriptive statistics and the Chi-Square test (MedCalc Software); p-value≤0.01 was considered significant. Ethics This study was approved by the Life and Medical Sciences Research Ethics Committee of the University of Latvia (document no. 71-35/50). The parents of the children signed informed consent forms before their inclusion in the study. RESULTS Participant characteristics In total, 91 healthy children (34 females and 57 males) aged 1–18 months (median, 10 months; CI, 5.0–11.0) were included. The participants were divided into two groups: 0–6 and 7–18 months of age. Most of the children were born in term with a normal birth weight (median, 3.5 kg; CI, 3.3–3.9) and birth height (mean, 53.4 cm; CI, 52.5–54.3). Most children (77% [71/91]) were born vaginally. Exclusive breastfeeding (EBF) up to 4 months of age was administered to 48.8% (42/86) of the children, and 12.8% (11/86) of the children received EBF only for the first few days of life. Mothers of 13 children received antibacterial therapy during pregnancy (mostly owing to urinary tract infection); 28 mothers reported previous antibacterial therapy during delivery, mostly due to cesarean-section (C-section) or positive group beta-hemolytic Streptococcus test results between the 36th and 38th gestational weeks. Antibacterial therapy was administered to 16 children in the first 6 months of life. The detailed characteristics of the participants are shown in Supplementary Table 2. Taxonomic structure and diversity Metagenomic sequencing analysis in total yielded 1.9 billion raw sequence reads, with an average of 21.7±4.4 million reads per sample. After quality control, on average 21.5±4.4 million high-quality non-human origin sequences per sample were obtained. Among the 100 most abundant taxonomic units, 94 belonged to Bacteria, four to Viruses, and two units to Archea (Supplementary Fig. 1). The most abundant bacterial species belonged to 12 phyla: the majority belonged to the Bacilllota, Bacteroidota, and Actinobacteria. Differences among the most abundant taxonomic units in the different age groups (1–6 months and 6–18 months of age) are shown in Fig. 1. The most abundant species depending on age: at 1-6 months and 6-18 months of age. At the species level, the most abundant representatives of the microbiome were Bifidobacterium longum, Bifidobacterium breve, and Escherichia coli (average relative abundance 10.6±15.4; 6.1±16.1, and 4.6±8.7) (Supplementary Fig. 1). In total, we were able to detect 17 Bifidobacterium species: the most abundant being B. longum and B. breve, followed by B. pseudocatenulatum (3.02±6.5), B. dentium (2.1±9.6), B. adolescentis (2.0±5.1), and B. bifidum (0.9±2.3) (Fig. 2). The most abundant viral community was the family Tospoviridae with the most abundant representatives being genus Triavirus, Pecentumvirus, Mosigvirus, and Gelderlandvirus. The relative abundance of Bifidobacterium species in infants up to 18 months of age. The mean Shannon index across all samples was 2.2. Pairwise comparisons showed that alpha diversity measurements did not differ significantly between delivery types (p>0.05). However, the number of observed species was higher in girls (p=0.044), children without EBF at the time of sample collection (p=0.005), and children whose parents did not report any allergies (p=0.008). The Shannon index was higher in children without EBF (p=0.004) and in children whose mothers had not received antibiotic therapy during delivery (p=0.015) (Fig. 3A). Sample diversity measurements. (A) Pairwise comparisons of alpha diversity measurements. (B) Pairwise comparisons of NMDS distances. CS: cesarean section, NA: not available, NMDS: non-metric multidimensional scaling. We identified significant differences in the distances between such parameters as the duration of total EBF (p=0.001), EBF at the time of sampling (p=0.002), parent-reported allergy (p=0.035), antibacterial treatment during the first 6 months (p=0.046), and place of living (p=0.042). However, no significant differences were found between parameters such as type of delivery, children, sex of children, or antibiotic consumption by the mother during delivery (p>0.05) (Fig. 3B). Environmental factors affecting microbiome composition Several perinatal and environmental factors affect gut microbiome composition. The entities that showed the most significant differences (p<0.005) in the median abundances for the studied factors are shown in the table (Table 1). Table 1The variations in taxonomic units depending on influencing factorsStudied factorTaxonomic unitp-valueDelivery: C-section↑species Kitasatospora spp. MMS16_BH0150.001↑species Nocardia asteroides0.001Previous maternal antibacterial therapy during delivery↑genus Alcyoneusvirus0.003↑species Coprococcus comes0.002↑species Kitasatospora spp. MMS16 BH0150.001↑species Mycoplasma anserisalpingitidis0.001Previous maternal antibacterial therapy during pregnancy↑genus Formosa0.001↑genus Micrococcus0.001↑genus Mycoplasmoides0.001↑genus Schizotequatrovirus<0.001↑species Actinomyces spp. Chiba1010.001↑species Allobaculum mucolyticum0.001↑species Candidatus Liberibacter asiaticus0.001↑species Clostridium baratii<0.001↑species Clostridium spp. SY8519<0.001↑species Clostridium taeniosporum<0.001↑species Cytobacillus kochii0.002↑species Dialister pneumosintes0.001↑species Granulicatella elegans0.001↑species Klebsiella grimontii0.002↑species Leptotrichia hongkongensis0.001↑species Metamycoplasma hominis0.001↑species Methanococcus aeolicus0.001↑species Methanococcus vannielii<0.001↑species Methanoplanus endosymbiosus0.001↑species Methanothermococcus okinawensis0.001↑species Oceanihabitans spp. IOP_320.001↑species Ornithinimicrobium ciconiae0.001↑species Pseudactinotalea spp. HY1580.001↑species Pseudoalteromonas issachenkonii0.001↑species Rheinheimera mangrovi0.001↑species Rothia nasimurium0.001↑species Silviavirus stau20.001↑species Singapore grouper iridovirus0.001↑species Spiroplasma kunkelii0.001↑species Vitreoscilla filiformis0.001Exclusive breastfeeding up to 6 months of age↓species Cardiobacterium hominis0.003↓species Endozoicomonas spp. 4G0.003↓species Kitasatospora spp. MMS16_BH015<0.001↓species Paraburkholderia caribensis<0.001↓species Streptococcus spp. 116_D40.003↓species Sulfurivermis fontis0.003Infants undergoing weaning↓species Clostridium butyricum0.001↓species Corynebacterium kroppenstedtii0.001↓species Corynebacterium tuberculostearicum0.0006↓species Helicobacter pylori0.001↓species Klebsiella michiganensis0.001↓species Ruthenibacterium lactatiformans0.001↑species Anaerostipes hadrus<0.0001↑species Blautia wexlerae<0.0001↑species Dorea longicatena0.003↑species Faecalibacterium prausnitzii<0.0001↑species Roseburia intestinalis0.001Living in a suburban region↑genus Novosphingonium0.001↑species Borrelia venezuelensis0.001↑species Desulfovibrio piger<0.001↑species Erwinia spp. QL_Z30.001↑species Manheimia haemolytica0.001↑species Pseudomonas spp. LS12120.001↑species Vagococcus hydrophili0.001 The variations in taxonomic units depending on influencing factors The greatest shift in taxonomic units was observed in children whose mothers received antibacterial therapy during pregnancy and/or delivery. Overall, significantly higher abundances of 30 different genera and species were observed in individuals who received prior maternal antibacterial therapy during pregnancy (Table 1). In particular, the prevalence of several environmental and potentially pathogenic bacteria was higher in infants whose mothers had received antibacterial therapy during pregnancy, as well as during the time of delivery. In cases of treatment with antibiotics during delivery, a higher average abundance of the genus Alcyoneusvirus (47.5 vs. 41.5, p=0.003), species Coprococcus comes (50.3 vs. 40.2, p=0.002), Kitasatospora MMS16 BH015 (48.7 vs. 41.0, p=0.001), and Mycoplasma anserisalpingitidisv (48.7 vs. 41.0, p=0.001) was observed. Furthermore, in cases of C-section delivery, species Kitasatospora MMS16 BH015 (51.8 vs. 43.2, p=0.001) and Nocardia asteroides (50.5 vs. 43.5, p<0.001) showed higher abundance compared to that with vaginal delivery. Previous maternal antibacterial treatment (during pregnancy and/or delivery) also had a significant impact on an infant’s fecal virome. Viruses detected in significantly higher abundance in children delivered by C-section belonged to Pimascovirales and Megavirales. The predominant viral entities in the infant gut virome, whose mothers had received antibiotic treatment, belonged to the bacteriophage order Caudovirales. The average abundance of the following species was in smaller amounts in children with EBF up to 6 months of age compared to that in children without EBF up to 6 months of age: Cardiobacterium hominis (42.5 vs. 48.8, p=0.002), Endozoicomonas spp. 4G (42.5 vs. 48.8, p=0.002), Kitasatospora MMS16 BH015 (42.0 vs. 49.8, p=0.0008), Paraburkholderia caribensis (42.0 vs. 49.8, p=0.0008), Streptococcus 116 D4 (42.5 vs. 48.8, p=0.002), and Sulfurivermis fontis (42.5 vs. 48.8, p=0.002). Furthermore, a higher average abundance of species Kitasatospora MMS16 BH015, which belongs to the phylum Actinomycetota, was observed in several groups: in infants delivered by C-section (53.7 vs. 43.2, p=0.001) and in cases of maternal antibacterial treatment during delivery (48.7 vs. 41.0, p=0.001); however, they were detected in smaller amounts in children with EBF up to 6 months of age (42.0 vs. 49.8, p=0.0008). The second largest shift was observed in infants after the start of weaning (Table 1). Higher average abundances were detected in bacterial species belonging to the phylum Bacillota and families Lachnospiracea and Ruminococcaceae. Multiple regression analysis (adjusted for previous antibacterial treatment, age group, type of delivery, and feeding type) revealed that the average relative abundance of species Kitasatospora MMS16 BH015 was significantly associated with the age group of 7–18 months (p=0.04) and previous maternal antibacterial treatment during delivery (p=0.017). In the total sample, the abundance of species Nocardia asteroides, adjusted for age, was associated with C-section (p=0.018). However, it was significantly associated with delivery type at the age of 0-6 months (p=0.04), but not with delivery type at the age of 7–18 months (p=0.15). In addition, the average relative abundance of several microbiome representatives was more common in children living in rural regions compared to that in those living in cities: genus Novosphingonium (46.5 vs. 40.0, p=0.001); and species Borrelia venezuelensi (46.5 vs. 40.0, p=0.001), Desulfovibrio piger (51.0 vs. 38.6, p=0.0008), Erwinia QLZ3 (47.7 vs. 39.6, p=0.001), Manheimia haemolytica (46.5 vs. 40.0, p=0.001), Pseudomonas LS1212 (46.5 vs. 40.0, p=0.001), and Vagococcus hydrophili (46.5 vs. 40.0, p=0.001), respectively. Antimicrobial resistance gene profile of infants A total of 302 AMR genes were identified within the gut microbiome of infants, of which the most frequently encountered genes were associated with glycopeptide antibiotic resistance: vanG (detected in 90 infants), vanT (detected in 89 infants), vanW (detected in 89 infants), vanY (detected in 87 infants), and adeF (detected in 79 infants) (Supplementary Table 3). Furthermore, significant differences were observed in the distribution of AMR genes, depending on several risk factors (type of delivery, antibiotic treatment during pregnancy, and delivery) (Table 2). Table 2The most frequently detected antimicrobial resistance genes depending on influencing factorsFactorThe amount of AMR genes (%) from a total sampleSignificantly more frequently detected AMR genesp-valuePrevious maternal antibacterial therapy during pregnancy62.6% (189/302)ANT 4 lb0.01patB0.006No previous maternal antibacterial therapy during pregnancy97.7% (295/302)vanH gene in vanD cluster0.01vanHgene in vanO cluster0.001vanXY gene in vanG cluster0.005vanY gene in vanA cluster0.006C-section delivery73.5% (222/302)aad 60.004APH_30.004FosA20.01Klebsiella pneumoniae KpnH0.01LEN130.006marA0.01Morganella morganii gyrB conferring resistance to fluoroquinolones0.010.008tetB P0.007vanR gene in vanD cluster0.01Vaginal delivery88.4% (267/302)mdtG0.01Previous maternal antibacterial therapy during delivery81.5% (246/302)aad 60.002APH_30.01baeR0.004marA0.01Morganella morganii gyrB conferring resistance to fluoroquinolones0.002No previous maternal antibacterial therapy during delivery86.1% (260/302)cblA10.01ErmF0.01AMR: antimicrobial resistance, ANT: aminoglycoside nucleotidyltransferase. The most frequently detected antimicrobial resistance genes depending on influencing factors AMR: antimicrobial resistance, ANT: aminoglycoside nucleotidyltransferase. DISCUSSION Initially, our understanding of the gut microbiome of infants was primarily gained through the employment of the 16S rRNA gene sequencing approach; however, recently, owing to the lowering of sequencing costs, novel knowledge has been accumulated using metagenomic shotgun sequencing analysis methods. The results of the present study showed that metagenomic sequencing of the infant fecal microbiome could provide deeper insights into the composition of the gut microbiome, emphasizing some unusual species that might be influenced by different environmental and perinatal factors. Recent data generated by 16S gene sequencing have shown that the intestinal microbiome of infants is dominated by bacteria from six phyla: Firmicutes, Bacteroidetes, Actinobacteria, Proteobacteria, Fusobacteria and Verrucomicrobia [19]. Our data show that microbes from the phyla Bacillota and Bacteroidetes are also a significant part of the infant microbiome; moreover, representatives of the genus Bifidobacterium dominate the infant microbiome. Similarly, Milani et al. [20] reported the dominance of Bifidobacterium in the guts of healthy breastfed infants, the most abundant being B. longum and B. breve. Breast milk enhances the predominance of Bifidobacterium species in the infant gut owing to the high concentration of human milk oligosaccharides (HMO). Therefore, the higher prevalence of B. longum could be explained by its role in HMO utilization. Similarly, Garrido et al. [21] reported that Bifidobacterium longum subsp. infantis showed homogeneous HMO-utilization patterns through its ability to effectively utilize diverse HMO species as the sole carbon source. Other strains of Bifidobacterium are more diverse in HMO utilization; for example, some strains of B. bifidum are unable to grow on several HMO species [21]. In addition, the relatively low abundance of B. bifidum could be explained by the inter-sample variety and the impact of EBF at the time of sample collection. According to our data, children who were breastfed at the time of fecal sample collection displayed a higher proportion of B. longum, while those who did not receive human milk were more often colonized by B. bifidum. Furthermore, the possible impact of concomitant probiotic treatment could influence the composition as over-the-counter probiotics often contain B. bifidum. Apart from the bacterial fraction of the gut microbiome, we also detected reads that originated from viruses, and the most abundant group was phages (according to annotations the 10 most common viruses were phages of Escherichia, Listeria, Salmonella, Bacillus, Klebsiella, Lactobacillus, and Pseudomonas). As previously demonstrated by 16S gene sequencing, the main factors influencing the gut microbiome are the type of delivery and feeding. Vaginally delivered infants had an increased abundance of Bacteroides, Bifidobacteria, and Lactobacillus in the first days and months of life, while in breastfed infants, a predominance of Bifidobacteria was observed [2223]. In addition, a history of recent antibacterial therapy in either the mother or infant could also greatly modify microbiome richness and composition [2425]. Similar tendencies were also demonstrated in our results, where the type of delivery as well as maternal antibacterial therapy within the perinatal period demonstrated a significant impact on the microbiome composition, thus emphasizing the importance of early-life environment in the colonization of the infant gut microbiome. Although the discussion on fetal gut sterility is still open, the potential effects of prenatal antibacterial treatment should be considered. Our data show an interesting shift in the taxonomic profile of the infant gut microbiome in cases where maternal treatment with antibiotics occurred during pregnancy. As previously reported, the maternal microbiome from different parts of the body, including the gut microbiome, is transmitted from mothers to their children within the first few hours after birth [26]. Furthermore, these changes might be a consequence of alterations in the mother’s intestinal microbiome, which contribute to the formation of the intestinal microbiota of vaginally born infants [27]. The impact of delivery mode on the infant gut microbiome has been studied more often in infants aged 0–6 months. For example, loss of Bacteroides colonization in C-section-delivered infants was detected especially in the first 2 weeks of life [28]. Furthermore, Reyman et al. [23], in a study of 120 children (46 born by C-section), observed differences in the early gut microbiome composition between vaginally delivered infants and those delivered by C-section, with a maximum effect of delivery mode after 1 week of life and at the age of 2 months. Similarly, Shao et al. [29] reported high colonization by opportunistic pathogens associated with the hospital environment, including Enterococcus, Enterobacter, and Klebsiella species in babies delivered by C-section. To our knowledge, no previous studies have reported higher numbers of the phylum Actinomycetota in C-section infants using deep sequencing, as shown in our study. Furthermore, it is unclear whether C-section itself or the use of antibiotics during C-section can cause changes in the infant gut microbiome [30]. This agrees with our finding that the abundance of species Kitasatospora sp. MMS16 BH015 was higher in infants born by C-section, as well as in cases of treatment with antibacterial therapy during delivery. However, multiple regression analysis showed that antibacterial therapy during delivery was independently associated with a higher abundance of Kitasatospora. Thus, Kitasatospora could be an indirect marker indicating that changes in the gut microbiome are more related to antibacterial therapy than to the cross-section itself. Although little is currently known about Kitasatospora, Li et al. [31] provided evidence supporting Kitasatospora as a separate genus belonging to the family Streptomycetaceae, thus showing high secondary metabolic potential. Feeding pattern is another important factor influencing the gut microbiome of infants. Interestingly, Species Kitasatospora MMS16 BH015 was observed in larger amounts, depending on the delivery mode (C-section) and antibacterial therapy (during delivery), but in smaller amounts in children with EBF, indicating that EBF could potentially reduce or change the amount of some taxa. Liu et al. [32] hypothesized that feeding patterns could influence the effect of delivery mode on the infant gut microbiome. They revealed no significant difference in the gut microbiota between vaginally delivered and C-section-delivered infants in the EBF groups, suggesting that EBF may alleviate the disturbance of the gut microbiota in infants delivered by C-section [32]. We observed species Streptococcus sp. 116 D4 in smaller amounts among breastfed infants. This is consistent with the data of Ma et al. [33], who showed that Streptococcus levels were significantly lower in breastfed infants at 40 days of age. Similar to our data, in a clinical study of infants during the transition, a significant increase in species belonging to Ruminococcaceae and Lachnospiraceae was detected after the introduction of solid foods [34]. Furthermore, species Blautia and Roseburia (belongs the family Lachnospiraceae) were reportedly some of the main producers of short-chain fatty acids in the human intestinal tract [35]. As reported previously, the abundance of butyrate-producing bacteria increases after the introduction of solid foods [36]. The process of establishing the gut microbiome is also affected by various environmental factors: one of those is the living condition and geographical location. We detected an increase in several bacterial species in infants living in suburban regions. Most belonged to the phylum Pseudomonadota, while others belonged to Spirochaetota and Deltaproteobacteria. All these bacteria are probably opportunistic bacteria within the human gastrointestinal tract. Nevertheless, based on the hygiene hypothesis, the gut composition differs among children living in rural regions. For example, Manheimia haemolytica is a significant causative agent of enzootic pneumonia, whereas Erwinia spp. are plant pathogens [3738]. Furthermore, geographical origin affects the overall microbial biodiversity of the infant gut microbiome [39]. To date, the taxonomic composition of the gut microbiome has been linked to the AMR load. Early life reflects a period of elevated resistome burden in the infant gut microbiome [40]. Reportedlyt, bacteria belonging to phylum Pseudomonadota and Gammaproteobacteria, especially Escherichia coli, are among the major carriers of AMRs [41]. In the present study, Pseudomonadota was one of the most abundant phyla, accounting for the 100 most abundant taxonomic units in all infants (Supplementary Fig. 1). Lebeaux et al. [42] estimated that infants born via C-section delivery had a higher risk of harboring unique ARGs and an increased risk for overall ARG relative abundance at the age of 1 year compared to those in infants born vaginally. However, observational evidence has shown that antibiotic-resistant bacteria colonizing the maternal vagina and intestines can be transferred to neonates during vaginal birth [43]. This idea is consistent with our data showing that vaginally born infants have an even higher number of AMR genes than do infants delivered by C-section. This finding emphasizes the importance of maternal antibacterial therapy during pregnancy, which may be associated with an increased resistome burden. Given the widespread wisdom that gut microbiome composition is largely influenced by factors such as the dietary patterns of different socioeconomic cultures, geographic regions, and environmental attributes, in the present study, we characterized Latvian infants using high-throughput metagenomic sequencing analysis. Although the obtained data may not be generalizable to the national level, this study provides important knowledge about the sensitivity of the infant gut microbiome to various medical and environmental determinants. In conclusion, metagenomic sequencing of the infant fecal microbiome, compared to 16S rRNA analysis, provides a broader insight into the composition of the gut microbiome, demonstrating the presence of some unusual species, thus emphasizing the influence of different perinatal and environmental factors. An infant's gut microbiome mainly consists of bacteria; Bifidobacterium species being the most prevalent taxonomic early colonizers in the infant gut microbiome. Characteristics of a viral fraction of the microbiome emphasize the possible effects of phages in modulating the bacterial composition of the gut microbiome. Antibacterial treatment during pregnancy and delivery may have a greater impact on the infant microbiome than was previously estimated. Rather, the high prevalence of ARGs (even in children without previous antibacterial therapy) demonstrates the importance of the gut microbiome in the spread of antibiotic resistance in early life, and the need for studies on possible modifications of the resistome to reduce the carriage of ARGs.