This study selected the Web of Science Core Collection (WOSCC) and Scopus as data sources, which cover high-quality scientific literature worldwide and are widely regarded as preferred tools for bibliometric analysis (Ding and Yang, 2020). An advanced search strategy was employed using the query: TS = (“bacteriophage” OR “phage therapy” OR “phage-host interaction” OR “virome”) AND (“gut microbiota” OR “intestinal flora” OR “gut microbiota” OR “gut flora”). The publication date range was restricted to January 1, 2012, to October 22, 2025, and only articles and review publications were included. The first author conducted the search and saved the data. The second and third authors independently screened the literature by reviewing titles and abstracts based on the inclusion/exclusion criteria, with disagreements resolved through consultation with the corresponding author. The exported metadata included titles, authors, keywords, citations, journals, institutions, and references, saved in plain text format (Chang et al., 2023). The study selection process was conducted in accordance with the PRISMA guidelines, and the inclusion and exclusion criteria are summarized in the PRISMA flowchart (Figure 1). In brief, records retrieved using predefined keywords were screened for relevance to bacteriophage regulation of the gut microbiome. Finally, CiteSpace, VOSviewer, and other bibliometric tools were used for multidimensional analysis and visualization of research outcomes. Since all data were obtained from public databases without involving human subjects or private information, ethics committee approval was not required.

The data analysis and visualization tools included CiteSpace (version 6.1.R6), VOSviewer (version 1.6.18), Scimago Graphica, and Microsoft Excel. Microsoft Excel was employed to calculate annual publication volumes by country/region and to generate line charts. CiteSpace and VOSviewer were used to visualize: (1) countries/regions and institutions; (2) authors and collaborative networks; (3) journal and reference co-citations; and (4) keyword co-occurrence clusters and bursts.

The visualization networks represent topological structures consisting of nodes and links. In these visualizations, node size corresponds to publication output, while link thickness indicates collaboration strength (Feng et al., 2023). VOSviewer provides three visualization modes: (1) network, (2) overlay, and (3) density visualizations, which collectively reveal cluster structures and identify pivotal nodes (Othman et al., 2022). Keyword co-occurrence analysis identified research hotspots within this field. In these visualizations, node size and font weight represent keyword frequency, whereas centrality values indicate conceptual importance. Nodes with centrality values ≥ 0.1 were classified as pivotal nodes. High-frequency, high-centrality keywords indicated established research priorities within the field. Keyword cluster analysis identified primary research directions. Lower cluster numbers corresponded to larger thematic groups. Cluster validity was assessed using Silhouette (S) and Modularity (Q) scores: S ≥ 0.5 indicated acceptable clustering; S ≥ 0.7 suggested strong reliability; and Q ≥ 0.3 denoted statistically significant cluster structures (Sabe et al., 2022; Isola et al., 2025). Temporal keyword emergence mapping identified evolving research frontiers. Emergence strength quantified intensity of sudden increases in keyword usage frequency, with higher values indicating more pronounced shifts. Journal Impact Factors were obtained from the Journal Citation Reports (2025 edition) in Web of Science.

To strengthen analysis validity, we normalized synonyms (e.g., “bacteriophage”/“bacteriophages,” “phage”/“bacteriophage,” “intestinal”/“gut”). For VOSviewer visualizations, we set display thresholds based on network density, including only items with connections. All other parameters used default settings. CiteSpace parameters included: analysis period (January 2012—October 2025) with yearly time slices. Pruning methods were “pathfinder” and “pruning sliced networks,” with all other parameters default.

Our search retrieved 729 records, with 687 articles and reviews meeting the inclusion criteria after screening.

Publication metrics served as key indicators of research progress and field development. We used Microsoft Excel to analyze the annual number of publications related to phage and gut microbiota research (Figure 2). Annual publications showed consistent growth from 2012 to 2025. Publications peaked in 2023 (n = 108 articles). The recent upward trend suggests growing academic interest and field maturation.

Sixty-three countries/regions contributed publications in this field, while Table 1 presents the top five most productive countries/regions. China produced the most publications (n = 174; 25.3%), followed by the United States (n = 156) and Italy (n = 36). These countries demonstrated both high productivity and central collaborative positions. Total Link Strength (TLS) analysis revealed the United States, United Kingdom, Germany, and China as key nodes in the collaboration network.

VOSviewer visualizations confirmed these patterns (Figures 3A–D). Thirty-one countries/regions published ≥ 5 papers each. The United States received the most citations, followed by China and the United Kingdom. The most frequent collaboration occurred between the United States and China, driving international knowledge exchange. This pattern shows how core nations disproportionately influence field advancement. However, research output remains unevenly distributed, with most publications originating from few countries. While promoting stable networks, this structure highlights the need for broader global participation. Such expansion would enhance resource sharing while reducing geographic bias and research duplication.

A total of 1,081 institutions published articles in the field of phage regulation of the gut microbiota, with the top five institutions listed in Table 2. The University of California, San Diego and the University of Copenhagen produced the highest number of publications (17 each). VOSviewer analysis revealed 43 institutions with five or more publications. As shown in Figure 4A, different colors represent distinct collaboration clusters, with most clusters exhibiting intra-national collaboration patterns. Visual knowledge maps of research institutions generated using Citespace (Figure 4B) reveal that publications in phage-mediated gut microbiota regulation predominantly originate from a small number of institutions, such as the University of California, San Diego and the University of Copenhagen. Overall, achievements in this field are primarily driven by comprehensive public research universities with medical programs and internationally renowned institutions. Leveraging their resource platforms and academic strengths, these institutions play a central role in advancing higher-quality research and enhancing its impact.

The included literature involved 3,842 authors, among whom 29 had published ≥ 5 papers (Figure 5). The co-occurrence network of authors generated by CiteSpace comprised 541 nodes and 1,257 connections, with a density of 0.0086. This low density indicates limited collaboration among authors, suggesting that a comprehensive large-scale collaborative network has yet to form. Five collaborative teams emerged, centered around Hill Colin, Schnabl, Bernd, Li, Shenghui, Xie, Mingxu, and Wang, Shumin. The most prolific author was Hill, Colin (15 papers), followed by Schnabl, Bernd (10 papers), Zhang, Yue (9 papers), and Li, Shenghui and Ross, R. Pau (both 8 papers) (Table 3). Author collaboration networks aim to reveal the most active and prolific authors and co-authors, visualize collaboration intensity, identify major collaborative teams and potential research partners within the field, and facilitate the establishment of closer collaborative networks (Xia et al., 2022). Figure 5 and Table 3 indicate that while numerous authors conduct research in this field, the overall collaboration structure remains fragmented, with relatively weak collaborative ties. This suggests a need to further strengthen cooperation and exchange between research teams, share clinical experience and research findings, and promote the development of high-quality research output in this field.

A total of 687 articles cited 35,689 references (Table 4). The paper by Jason M Norman in 2015 received the highest number of citations (140), indicating its exceptional research value and influence in the field of phage regulation of the gut microbiota. Using VOSviewer for co-citation analysis, we set the minimum number of co-cited references to 20. This identified 133 articles meeting this threshold, forming three major clusters (Figure 6A). CiteSpace performed reference clustering analysis, identifying 11 major subtopics relevant to phage regulation of the gut microbiome. The Modularity Q-value was 0.7435 (> 0.3) and the Silhouette S-value was 0.8694 (> 0.7), indicating that the co-citation clusters were reliable and significant, with stable and dependable clustering results (Figure 6B). Citation emergence analysis (Figure 6C) revealed that Jason M Norman’s 2015 paper “Disease-specific alterations in the enteric virome in inflammatory bowel disease” exhibited the highest emergence intensity (20.25), reflecting its significance within the field of phage-mediated gut microbiota regulation. Through multi-cohort metagenomic sequencing with cross-regional validation, combined with virus-like particle (VLP) enrichment and 16S rRNA analysis, this study systematically revealed the dynamic changes in the viral microbiome associated with inflammatory bowel disease (IBD) and its impact on host pathophysiology. The study not only provides crucial evidence for assessing viral risks in inflammation and dysbiosis but also lays the groundwork for developing viral diagnostic tools and phage-targeted therapies, advancing research into the interaction mechanisms between the gut virusome and microbiome.

The 687 articles in this field were published across 273 journals, with 20 journals publishing ≥ 5 articles. Frontiers in Microbiology published the most articles (n = 36), followed by the International Journal of Molecular Sciences (n = 20) and Microorganisms (n = 17). Figures 7A,B and Table 5 present the top 10 journals by publication volume and citation impact. Dual-map overlay analysis can be used to examine the distribution characteristics of disciplines and journals and reveal connections between them. The left side represents the disciplinary distribution of citing journals, while the right side shows the disciplinary distribution of cited journals. Paths of different colors indicate citation relationships between them (Ma et al., 2021; Jiang et al., 2025), with two paths representing the disciplinary transition links between citing and cited journals. Dual-map overlay analysis indicates that citing literature related to phage regulation of gut microbiota predominantly appears in journals of Molecular/Biology/Immunology, Medicine/Medical/Clinical, Dentistry/Dermatology/Surgery, and Neurology/Sports/Ophthalmology, while cited literature primarily originates from Molecular/Biology/Genetics journals. This pattern suggests phage-gut microbiota research emerged from fundamental life science advances. Due to its immense application potential, it rapidly attracted participation from various clinical medical disciplines. This demonstrates that the field’s development relies on deep integration between disciplines such as molecular biology, genetics, microbiology, and bioinformatics with clinical medicine and various clinical specialties, ultimately forming a highly interdisciplinary research paradigm. Analysis reveals that journals in this field exhibit high impact factors and quality standards, ensuring the rigor and academic integrity of published research.

Keywords encapsulate a study’s core concepts and thematic focus. Systematic keyword analysis identifies research themes through high-frequency terms, revealing field developments and emerging directions (Fu et al., 2023). Keyword co-occurrence mapping visually classifies research themes, highlighting current foci and trends. The keyword co-occurrence map is shown in Figure 8A. High-frequency keywords comprised: phage therapy, gut virome, inflammatory bowel disease, diversity, and bacterial infection. The top 10 keywords by frequency are listed in Table 6. Key nodes in this study include Children (0.14), Therapy (0.11), Immune response (0.10), Antimicrobial resistance (0.10), Community (0.10), and Colonization (0.10). The keyword clustering network diagram is shown in Figure 8B. Results indicate a literature clustering S-value of 0.7776 and a clustering Q-value of 0.5186, confirming the clustering as statistically significant and reliable. The 13 clusters reflected major research trajectories and temporal evolution. Further summarizing each cluster’s research focus, the 13 clusters were consolidated into four major categories, as shown in Table 7. 1. Virome and metagenomic innovations: This category represents technology-driven frontier exploration, focusing on revealing the diversity, community structure, and functions of the gut virome through novel techniques such as viral metagenomics and high-throughput sequencing. It aims to unravel the “viral dark matter” and provide a methodological foundation for understanding phage-host interactions. 2. Resistance evolution and host-microbe interactions: This category focuses on molecular and ecological mechanisms, emphasizing bacterial adaptive evolution under environmental pressures like antibiotics and host immunity. Examples include the “arms race” between CRISPR-Cas systems and anti-CRISPR proteins, and the role of microbial metabolites such as short-chain fatty acids (SCFAs) in regulating host signaling pathways and outcomes of bacterial infections. 3. From phage therapy to microecological intervention: This category signifies a paradigm shift from single-pathogen eradication to holistic microecological regulation. Core strategies include targeted bacterial killing using natural or engineered phages, alongside microecological interventions like fecal virus community transplantation to restore gut microbial balance for disease treatment. 4. Immunomodulation through microbiome engineering: This category focuses on host-level mechanisms and applications, delving into how dysbiosis triggers chronic inflammatory diseases by disrupting the intestinal barrier, altering immunomodulatory molecules like toll-like receptors, and affecting SCFAs. It also evaluates intervention strategies such as Fecal Microbiota Transplantation (FMT) in improving disease outcomes by restoring immune homeostasis. Figure 8D presents keyword salience analysis. “Sequence” showed highest salience (5.48), with temporal analysis revealing two phases: Phase I (2012–2019) emphasized: pediatric studies, C. difficile, viral ecology, and colonization resistance. Key themes included gut disease and pathogen studies, community structure and diversity analysis, viral ecology exploration, and colonization and resistance mechanisms. Phase II (2020-present) shifted toward: inflammatory mechanisms, E. coli/P. aeruginosa pathogenesis, and virome functional dynamics. Building upon studies of community ecology and resistance mechanisms, this phase expands into inflammation and immune regulation, metabolic health and host interactions, and functional remodeling of the gut virome. This shift reflects an evolution in research focus from structural characteristics toward functional mechanisms and clinical translation. Keywords such as Gut virome, Disease, and Expression have emerged as prominent current research hotspots.

Since the early 20th century, epidemiological transitions have shifted global mortality patterns from infectious to non-communicable diseases (GBD 2019 Diseases and Injuries Collaborators, 2020). Consequently, host-microbiota interactions have emerged as a key research focus in biomedicine. Mounting evidence links gut dysbiosis to chronic diseases (cardiovascular, IBD, neurological, and allergic disorders), representing a global health challenge (Kayama et al., 2020). However, phage roles in microbial networks and host health regulation remain underexplored (Avellaneda-Franco et al., 2024).

Our analysis included 687 English publications (2012–2025). Publication volume increased from 2 papers in 2012 to 108 in 2023, showing a sustained upward trend with particularly notable growth during critical junctures like pandemic outbreaks. This indicates that as global public health events become more frequent, research on the association between phages and gut microbiota is gaining increasing attention from the global academic community. Its importance in fields such as health and disease prevention and control is continuously rising, suggesting that the practical demand for prevention and control drives the deepening of research. From a national perspective, China produced the highest number of publications, whereas the United States exhibited the strongest total link strength, indicating a more extensive international collaboration network. Notably, Western countries such as the United Kingdom, the United States, and Italy showed relatively higher average citations per publication, reflecting greater research impact. These patterns may be attributed to well-established research infrastructure and long-term investment in microbiome studies, earlier entry into the field allowing more time for citation accumulation, and higher international visibility resulting from English-language publications in high-impact or open-access journals (Zhou et al., 2020; Ryan et al., 2021). Institutions such as the University of California, San Diego and the University of Copenhagen have played a pivotal role in advancing global microbiome research through their outstanding contributions to understanding the mechanisms underlying gut microbiota dysbiosis, metabolic diseases, and phage interactions (Tanase et al., 2020; Zaky et al., 2021). However, geographic concentration and collaboration fragmentation remain challenges, requiring strengthened international cooperation. Future efforts should foster transnational scientific cooperation and multidisciplinary integration to establish a more systematic and collaborative global research network in this field. Collaboration networks revealed influential research teams led by Colin Hill, Bernd Schnabl, and R. Paul Ross. Colin Hill’s team spans diverse research domains, from phage structural biology to clinical microbiome interventions (King, 2024). His work pioneered the atomic-level structure of crAssvirus—the most abundant phage in the human gut—via cryo-electron microscopy, providing foundational theoretical insights into phage assembly and infection mechanisms (Shkoporov et al., 2018; Bayfield et al., 2023). His team pioneered standards for microbial therapeutics and postbiotics while translating microbiome research into clinical practice (O’Toole et al., 2017; Salminen et al., 2021; Mosca et al., 2022). Colin Hill’s research centers on the dynamic symbiotic relationship between phages and host bacteria. For instance, his research proposed the “phage cocktail therapy” strategy and demonstrated its ability to significantly reduce pathogenic bacterial abundance in vitro models (Buttimer et al., 2022; Cortés-Martín et al., 2025). Additionally, he introduced the concept of fecal viral transfer (FVT), proposing that phages play a crucial role in gut microbiota restoration following antibiotic interventions. This framework facilitates the development of non-virographic formulations and offers novel insights (Draper et al., 2020). Furthermore, the team elucidated the mechanisms linking phages to metabolic diseases and IBD, emphasizing synergistic strategies combining narrow-spectrum antibiotics with probiotics to counter the spread of resistance genes caused by antibiotic overuse (Murphy et al., 2013; Clooney et al., 2019; O’Connor et al., 2025). These studies not only shape the knowledge structure of this field but also advance the understanding of phage-gut microbiota-host interactions, providing a theoretical foundation for precision microbiome interventions and personalized therapies. Key findings appear predominantly in high-impact journals like Frontiers in Microbiology and International Journal of Molecular Sciences. This interdisciplinary research bridges microbiology, chemistry, and biomedicine, focusing on mechanistic insights and clinical translation (Mei et al., 2022; Zhou et al., 2023). Most journals feature high impact factors, with JCR rankings predominantly in Quartiles 1 and 2, reflecting the field’s overall high research quality and steadily increasing academic influence.

Bibliometrics systematically analyzes publication data to reveal research trends and patterns (Li et al., 2024). Keyword analysis (co-occurrence, clustering, emergence) identifies research themes, while co-citation networks map intellectual foundations Thus, keyword analysis in phage-microbiota research helps forecast emerging directions (Chen, 2017; Wang J. H. et al., 2024). Our analysis identified four research hotspots: innovations in virology and metagenomics technologies, evolution of drug resistance and microbiota-immune interactions, progression from phage therapy to microbiome modulation, and immune regulation and microbiome interventions.

While this study offers valuable insights into phage-microbiome research trends, some limitations should be noted: First, using only WoSCC and Scopus databases may have excluded relevant studies from PubMed and other sources. This choice was made to ensure standardized citation metadata and compatibility with bibliometric analysis tools, while minimizing potential duplication arising from overlapping or translated records. Second, the English-only inclusion criterion may introduce language bias by excluding non-English publications. This limited data scope may inadequately represent global research on phage-microbiome regulation. Lastly, the bibliometric visualization software used in this study does not distinguish authorship positions, such as first or corresponding authors, but ranks authors collectively based on publication output and co-authorship relationships. As a result, the authors identified as prolific in this study reflect overall research productivity rather than specific authorship roles or individual research influence.