Skip to main content
Download PDF

Network analyses unraveled the complex interactions in the rumen microbiota associated with methane emission in dairy cattle

Animal Microbiome volume 7, Article number: 24 (2025) Cite this article

Abstract

Background

Methane emissions from livestock, particularly from dairy cattle, represent a significant source of greenhouse gas, contributing to the global climate crisis. Understanding the complex interactions within the rumen microbiota that influence methane emissions is crucial for developing effective mitigation strategies.

Results

This study employed Weighted Gene Co-expression Network Analysis to investigate the complex interactions within the rumen microbiota that influence methane emissions. By integrating extensive rumen microbiota sequencing data with precise methane emission measurements in 750 Holstein dairy cattle, our research identified distinct microbial communities and their associations with methane production. Key findings revealed that the blue module from network analysis was significantly correlated (0.45) with methane emissions. In this module, taxa included the genera Prevotella and Methanobrevibactor, along with species such as Prevotella brevis, Prevotella ruminicola, Prevotella baroniae, Prevotella bryantii, Lachnobacterium bovis, and Methanomassiliicoccus luminyensis are the key components to drive the complex networks. However, the absence of metagenomics sequencing is difficult to reveal the deeper taxa level and functional profiles.

Conclusions

The application of Weighted Gene Co-expression Network Analysis provided a comprehensive understanding of the microbiota-methane emission relationship, serving as an innovative approach for microbiota-phenotype association studies in cattle. Our findings underscore the importance of microbiota-trait and microbiota-microbiota associations related to methane emission in dairy cattle, contributing to a systematic understanding of methane production in cattle. This research offers key information on microbial management for mitigating environmental impact on the cattle population.

Background

Methane (CH4), a greenhouse gas (GHG), significantly contributes to global warming [1]. The livestock sector represents 14.5% of human-induced GHG emissions, where feed production/processing and enteric fermentation from ruminants are the two main sources representing 45% and 39% of emissions in the livestock sector, respectively [2]. Among all ruminant species, dairy cattle accounts for 30.1% of the sector’s emission, and the emission in dairy cattle is mainly from ruminal CH4 emission [3]. As one of the major contributors of GHG emissions, dairy sector plays a key role in realizing Agriculture Net Zero through mitigating enteric CH4 emission [4]. Moreover, facing the high demand of milk and meat supply, CH4 emissions from livestock is predicted to increase continuously by 2050 [5]. These highlight the pressing need for innovative strategies to mitigate CH4 emissions to address environmental concerns while maintaining sustainable dairy production [6,7,8,9].

The rumen, largest digestive organ in ruminants, serves as an anaerobic ecosystem that fosters the growth of diverse microorganisms, including bacteria, archaea, fungi, and ciliate protozoa, which collectively influence CH4 production [10, 11]. Major factors that affect CH4 emissions from dairy cattle rumen include genetics [9, 12], management [13], feeding strategies [14], and lactation stages [15]. Ruminal methanogenesis, a key contributor of CH4 emissions [16], contains complex biological processes with diverse metabolic pathways [17,18,19]. Methanobrevibacter and Methanomassiliicoccus are the dominant rumen archaeal genera responsible for a significant portion of CH4 emissions [16]. Most rumen methanogens have hydrogenotrophic metabolisms, meaning they use electrons from H2 to reduce CO2 to CH4, an efficient way to reduce H2 concentrations in the rumen [20]. Meanwhile, ciliate protozoa from a vital component of the rumen microbial ecosystem, engaging in diverse metabolic pathways. Many of these functions arise from their close interactions with associated prokaryotic communities. For example, CH4 production can be intensified through interspecies hydrogen transfer between protozoa and archaea [21]. Together with a diverse bacterial community that ferments and degrades complex carbohydrates and proteins, they form a complex network that ultimately determines CH4 emissions. The effort to understand the complex microbial network is the key to developing mitigation strategies for ruminant CH4 production [22,23,24].

Despite previous research on complex metabolic networks in rumen [25], a gap remains in the understanding of the interactions within rumen microbiota on CH4 emissions. Previous studies have often focused on either the taxonomic composition of the microbiota or the host's genetic variations [12, 26,27,28]. The application of advanced system biological tools, such as Weighted Gene Co-expression Network Analysis (WGCNA) [29], to investigate the comprehensive relationship between the microbiota and CH4 emissions remains unexplored. This study utilizes WGCNA to study the interactions within rumen microbiota and its link to CH4 emissions in dairy cattle [30, 31]. WGCNA enables the identification of the microbial interactions associated with traits by generating modules [29]. In this study, we hypothesized that specific rumen microbial community as one module could be used to explain the CH4 emissions in dairy cattle. To address this hypothesis, we integrate CH4 emissions data and rumen 16S rRNA gene amplicon sequencing data from 750 Holstein cows to identify the microbial community structures and interactions that are associated with CH4 emission in this study.

Results

Rumen microbiota compositions and structures

We obtained 191,035,652 and 115,632,617 total reads for bacteria (mean ± SD: 226,077.695 ± 209,835.229) and archaea (mean ± SD: 106,843.116 ± 90,783.265), respectively (Supplementary Fig. 1). Additionally, a total of 59,445 and 1805 observations (ASVs) for bacteria and archaea were retained as the raw data for downstream analysis (750 cows). The alpha diversity analysis showed richness (Chao1, richness) and evenness (Shannon 2, Simpson) metrics for archaeal and bacterial ASVs (Supplementary Fig. 1). Compared to archaeal communities, bacterial communities performed a better richness and evenness (P < 0.001) in Supplementary Fig. 1. A total of 35 and 1393 ASVs remained for archaea and bacteria in the final datasets after quality control (relative abundance > 0.01% among half samples) [12]. The diversity and distribution of these microbial communities was shown in Principal Component Analysis (PCA) plots confirmed with PERMANOVA, which shown herds influenced sample clustering, while lactations appear to distribute randomly across both archaeal and bacterial samples (Supplementary Fig. 2).

In terms of taxonomic classification, two predominant archaeal classes within the Euryarchaeota phylum — Methanobacteria and Thermoplasmata, identified (Fig. 1A). The Methanobacteriaceae family, which comprise the genera Methanosphaera, and Methanobrevibacter, was distinctly represented in the phylogenetic branches (Fig. 1A). In the bacterial domain, the Bacteroidetes phylum, enriched by genus Prevotella, and their species Prevotella brevis, and Prevotella ruminicola (Fig. 1C).

Fig. 1
figure 1

Composition and abundance of rumen microbiota in cattle. Archaeal (A, B) and bacterial (C, D) community composition were annotated to species, genus, family, order, class, phylum, and domain levels. Node color represents the counts of ASV taxonomic annotation. Relative abundance of archaea (C) and bacteria (D) were summarized at genus level, respectively. NA means the non-annotations at genus level for bacterial communities in Fig. 1D

Full size image

The relative abundance of archaea highlighted 5 annotated genera, which were dominated by Methanobrevibacter and Methanomassiliicoccus (Fig. 1B). At the species level, there were 3 annotated species, including Methanobrevibacter millerae, Methanosphaera stadtmanae, and Methanomassiliicoccus luminyensis (Supplementary Fig. 3A). Bacterial community showed that the genus Prevotella was predominant. Other notable genera, like Bacteroides, Fibrobacter, Ruminococcus and Ruminobacter were also identified among the top abundant ASVs respectively (Fig. 1D). At the species level, dominant bacterial species included Prevotella brevis, Prevotella maculosa, Prevotella ruminicola, and Alistipes shahii, along with Bacteroides massiliensis, Bacteroides acidifaciens, and Bacteroides coprocola (Supplementary Fig. 3B).

The WCNA clusters associated with CH4 emission

The Module blue (MEblue) was notably correlated with CH4 was notable, exceeding 0.45 with a significant P-value (7e-37) (Fig. 2), including genera Methanobrevibacter, Prevotella, and Aminiphilus; species Prevotella brevis, Prevotella stercorea, Parabacteroides distasonis, Bacteroides massiliensis, SR1 genera incertae sedis, and Methanomassiliicoccus luminyensis (Fig. 3). Additionally, Module brown (MEbrown) exhibited a relatively high correlation with the Herd (r = 0.43, p = 4e-34) (Fig. 2), with dominant taxa including Prevotella, Rufibacter, and species such as Prevotella brevis, Prevotella albensis, Bacteroides coprocola, Parabacteroides distasonis, and Butyrivibrio proteoclasticus (Supplementary Fig. 5).

Fig. 2
figure 2

Modules-traits relationships of merged rumen microbiota community. Herd: The distinct groups of animals sampled from different farms. Lact (Lactation): The lactation number for each cow. DIM (Days in Milk): The number of days a cow has been in milk production during the sampling period. CH4 (Methane emissions): The corrected methane emissions for each cow per day. The P_values of correlation estimates are shown in brackets. The color of the rocks indicates the value of the correlation estimates

Full size image
Fig. 3
figure 3

WGCNA modules containing microbiota related to methane emissions with taxonomic annotations at highest level. These taxa were visualized alongside archaeal and bacterial species and genera nodes that significantly correlated to methane emissions. Nodes’ color represents their modules generated by WGCNA. Node’s shape represents the toNodes and fromNodes

Full size image

The merged ASV dataset (archaea and bacteria) remained with 722 samples and 1428 ASVs. The lactation numbers of each cow (Lact), days in milk (DIM), and farms (Herd) were correlated with ASVs to dissect the effect of the cow’s physiological status or the farm management. CH4 emissions were corrected with ASVs to reveal the effect of rumen microbiota on the CH4 emission. The resulting module dendrograms displayed the clustering of ASVs, identifying 6 modules for ASVs (Supplementary Fig. 4). Each module represents a group of highly interconnected microbial taxa, revealing co-abundance relationships and their links to CH4 emissions. The grey module was excluded due to unclassified co-expression patterns. The heatmap (Fig. 2) displays the correlations between each module and trait combination, emphasizing the strong links between MEblue-CH4 emissions and MEbrown-Herd. In contrast, Lact and DIM showed no significant module correlations (r < 0.3), suggesting that CH4-associated microbial dynamics are more influenced by host-independent factors like diet and farm conditions rather than individual physiological stages.

The Hub ASVs and their interactions

To gain deeper insight into the key species in the identified modules (Fig. 2), we identified hub ASVs in significant modules. In MEblue, 56 hub ASVs were identified as hub ASVs (Fig. 3), including two archaeal ASVs (arc asv8-arc asv15) annotated to Methanobrevibacter and Methanomassiliicoccus luminyensis. Most hub ASVs were annotated to species within Prevotella (Fig. 3). Notably, 17 microbial interactions were identified in MEblue, with the most complex network involving 18 ASVs (asv539-asv334-asv930-asv1004-asv133-asv118-asv16-asv787-asv500-asv231-asv644-asv517-asv577-asv1139-asv156-asv307-asv1675-asv1184), which belonged to the species Prevotella brevis, Prevotella ruminicol, Prevotella ruminicola, and Prevotella baroniae (Fig. 3). There were 12 two microbiota interactions, 3 three microbiota interactions and 1 five microbiota interactions. These simple interactions included Prevotella brevis-Prevotella ruminicola-Prevotella baroniae, Prevotella brevis-Lachnobacterium bovis, Prevotella ruminicola-Prevotella bryantii, and Prevotella ruminicola-Prevotella brevis-Barnesiella viscericola, as well as Paraprevotella xylaniphila within MEblue (Fig. 3).

Similarly, MEbrown interactions, driven by the high correlation between Herd (r = 0.43, p = 4e-34) (Fig. 2), including 33 significant ASVs clustered into seven interactions. These interactions prominently involved genera Prevotella, Rufibacter and species such as Prevotella brevis, Prevotella ruminicola Parabacteroides distasonis, Parabacteroides merdae, Bacteroides coprocola, and Flavonlfractor plautii (Supplementary Fig. 5).

Pathways annotation and differential functions of archaea and bacteria

There were 363 archaeal and 312 bacterial level 3 pathways (Supplementary Table 2), respectively. The top 30 abundances of level 3 pathways were further annotated to "Methane metabolism", "Transporters", “Ribosome”, and “Transfer RNA biogenesis” being the most notable in archaea as depicted in Supplementary Fig. 6A. Similarly, top abundant bacterial pathways were dominated by "Transporters", "DNA repair and recombination proteins", "Transfer RNA biogenesis”, and “Ribosome” (Supplementary Fig. 6B).

For the MEblue and MEbrown modules, there were 314 and 302 level 3 KEGG pathways (Supplementary Table 3). The top abundant pathways of MEblue and MEbrown were similar, which was dominated by "Transporters", “Ribosome”, "DNA repair and recombination proteins", and "Transfer RNA biogenesis” (Fig. 4A, B). Notably, in MEblue, “Methane metabolism” was observed in the top 30 abundant pathways (Fig. 4A). These two modules were enriched in the level 2 pathways “Protein families: genetics information processing”, “Amino acid metabolism” and “Carbohydrate metabolism”.

Fig. 4
figure 4

Level 3 KEGG functional pathways of WGCNA high correlated Modules (Blue [A]; Brown [B]) and hub ASVs in module blue (MEblue) (C) and module brown (MEbrown) (D) characterized by Level 2 pathways. This figure displays the top 30 abundant KEGG pathways. To enhance clarity, the bar length means the relative abundance of KEGG pathways at level 3. The different colors of bar represent the different level 2 pathways, which are described in the grey box on the right side of the figure

Full size image

There were 264 KEGG level 3 pathways for hub ASVs in MEblue (Supplementary Table 4). The top abundant pathways were “Transporters”, “DNA repair and recombination proteins”, “Transfer RNA biogenesis”, and “Ribosome”, which was similar to bacterial community (Fig. 4C). Interestingly, “Methane Metabolism” was identified in the top 30 abundant pathways, belonging to “Energy metabolism” (Fig. 4C), which was observed in MEblue as well (Fig. 4A). Meanwhile, 265 KEGG level 3 pathways were annotated for hub ASVs in MEbrown (Supplementary Table 4). The top abundant pathways were similar to MEblue and MEblue KEGG pathways, including “Transporters”, “DNA repair and recombination proteins”, “Transfer RNA biogenesis”, and “Ribosome” (Fig. 4D). The top abundant level 3 KEGG pathways of MEblue and MEbrown were all enriched in level 2 pathways “Protein families: genetics information processing”, “Amino acid metabolism”, and “Carbohydrate metabolism” (Fig. 4C, D).

Discussion

Based on the WGCNA results, MEblue was the most significant module, correlating moderately with CH4 emissions (r = 0.45, p = 7e-37) (Fig. 2). This module includes both methanogenic archaea (Methanobrevibacter and Methanomassiliicoccus luminyensis) and carbohydrate-fermenting bacteria (Prevotella brevis, Prevotella ruminicol, Prevotella ruminicola, and Prevotella baroniae) (Fig. 3), forming a metabolic network that contribute to CH4 emissions. Methanobrevibacter and Methanomassiliicoccus luminyensis are well-known for their roles in methanogenesis [32]. Both Methanobrevibacter and Methanomassiliicoccus luminyensis are methanogenic archaea that contribute to CH4 production in rumen, utilizing hydrogen and CO2 (hydrogenotrophy) or methylated compounds (methylotrophy) to produce CH4 [33, 34]. Notably, Methanomassiliicoccus luminyensis specially requires hydrogen as an electron donor, reducing methanol, methylamines into CH4 [35]. The co-occurrence of Methanobrevibacter and Methanomassiliicoccus luminyensis within MEblue suggests functional interactions, where Methanobrevibacter helps maintain low hydrogen partial pressure in the rumen, indirectly supporting Methanomassiliicoccus by creating favorable conditions for its methylotrophic methanogenesis [36].

In addition to archaea, MEblue also includes bacteria from genus Prevotella, particularly networks among Prevotella brevis-Prevotella ruminicola-Prevotella baroniae, Prevotella brevis-Lachnobacterium bovis, Prevotella ruminicola-Prevotella bryantii, and Prevotella ruminicola-Prevotella brevis-Barnesiella viscericola, as well as Paraprevotella-xylaniphila within MEblue (Fig. 3). The interactions between Prevotella brevis, Prevotella ruminicola, and Prevotella baroniae suggest synergistic relationships that enhance polysaccharide breakdown and hydrogen production, both of which are crucial for CH4 production [37]. These bacteria was highly abundant and co-clustered, emphasizing their role in carbohydrate fermentation, converting into short-chain fatty acids (SCFAs) [38, 39]. Among SCFAs, acetate indirectly contributes to CH4 production by methanogenic archaea. While acetate can serve as a substrate for methanogenesis in specific contexts, though its contribution to ruminal methane production is minimal due to the dominance of hydrogenotrophic and methylotrophic pathways and the rapid rumen passage rate, which limits acetogenic methanogens [39]. Furthermore, the cross-phylum microbial interaction (Prevotella brevis-Lachnobacterium bovis) is a suggests a synergistic metabolic network, where fermentative bacteria generate hydrogen as a byproduct, which is then utilized by methanogenic archaea for CH4 production. Additionally, Lachnobacterium bovis, which was found to interact with Prevotella brevis, produces intermediates like lactate and acetate, further enhancing hydrogenotrophic pathways that contribute to CH4 emissions [19]. Our results align with previous studies [17, 38] but provide more detailed insights into microbial interactions at the species level. Furthermore, these findings reinforce that MEblue represents a functionally cohesive module linked to CH₄ emissions through hydrogen-mediated microbial interactions.

While MEblue is significantly associated with CH4, other microbial modules also influence methane metabolism under different environmental conditions. For instance, MEbrown, another major module, contains similar fermentative bacteria (Prevotella brevis and Prevotella ruminicola), which are well-known contributors to carbohydrate fermentation and SCFAs production [39]. However, MEbrown networks exhibited greater diversity in bacterial taxa, including Flavonifractor plautii, Parabacteroides merdae, Barnesiella viscericola, and Alistipes putredinis (Supplementary Fig. 5). Unlike MEblue, which is moderately linked to CH4 emissions, MEbrown is more influenced by herd effects (r = 0.43, p = 4e-34) (Fig. 3), suggesting that herd-specific factors such as feeding practices and diet composition [40]. Also, Flavonifractor plautii contributes to flavonoid degradation [41], influencing microbial dynamics and contributing to SCFAs production. PCA analysis (Supplementary Fig. 2) further revealed that the herd-specific factors significantly shaped microbial community structure, with lactation effects being less pronounced. This may be partly due to the fact that our study was conducted in commercial dairy cattle farms where cattle were reared with potentially different management conditions. Variations in feed composition, fiber content, and nutritional balance across farms likely contributed to the microbial diversity within MEbrown. This underscores the importance of dietary and environmental factors in modulating microbial networks and their roles in CH4 emissions [40]. Understanding these environmental influences can help develop targeted intervention strategies to manipulate microbial communities for CH4 mitigation.

Since MEblue represents a microbial network closely associated with CH4 production, dietary and management interventions targeting this module could be effective in reducing methane emissions. Recent studies reported that strategies such as dietary modifications, probiotics, and feeding additives can affect the CH4 emissions by altering microbial community structure and metabolic pathways [42,43,44]. Probiotics can modulate gastrointestinal microbial. Their colonization in rumen improves feed efficiency, potentially reducing CH4 emissions [23]. Similarly, dietary interventions, including high-lipid diets [45], nitrate supplementation [46], and plant secondary metabolites (e.g., tannins and saponins) [47, 48], have been explored as effective CH4 mitigation strategies. Lipid supplementation can suppress methanogens by reducing hydrogen availability, while tannins can directly inhibit methanogens activity [45]. Furthermore, dietary nitrate supplementation provides an alternative hydrogen sink, outcompeting methanogenesis and reducing CH4 emissions [46]. Beyond diet, farm management practices such as precision feeding, controlled grazing, and strategic supplementation can also influence microbial communities. Precision feeding strategies that optimize fiber and protein balance can reduce hydrogen accumulation and CH4 formation [49]. Future studies should investigate the long-term effects of such interventions on microbial networks and rumen functionality.

Functional annotation of archaeal microbiota observed pathways associated with CH4 production, which was enriched in “Methane metabolism”. This result provides an insight into the direct role of archaea in the rumen ecosystem. In contrast, bacterial ASVs exhibited different functional structures, including "Transporters", "Ribosome", and "DNA repair and recombination proteins". These differences reflect the complementary roles of bacteria and archaea in the rumen, where bacteria contribute to the breakdown and fermentation of complex carbohydrates, providing precursors like hydrogen and SCFAs for archaeal methanogens [17, 34, 39]. Interestingly, despite the variability in taxonomic composition, the rumen microbiota's functional pathways appear conserved. Both the general bacterial ASVs, the WGCNA MEblue or MEbrown ASVs, and their hub ASVs were enriched in KEGG level 3 pathways "Transporters", "Ribosome", and "DNA repair and recombination proteins", which belong to level 2 pathways "Translators", "Membrane transport", and “Replication and repair” (Fig. 4, Supplementary Fig. 6). Our functional prediction results based on bacterial ASVs, WGCNA MEblue and MEbrown ASVs were similar. This similarity suggests functional redundancy among different bacterial taxa in the rumen, ensuring the stability and efficiency of microbial processes essential for host energy metabolism [16, 27, 50, 51]. Functional redundancy is a well-documented phenomenon in microbial communities and is thought to arise from environmental selection for critical biochemical processes, as observed in a recent study on cross-biome microbial networks [52]. To elucidate the specific roles of these interactions in CH4 production, future studies could employ metagenomics or metatranscriptomics to identify active metabolic pathways and their gene-level regulation within CH4 emissions.

Unlike SparCC [53], which focuses on pairwise correlations, WGCNA enables the identification of modules of highly correlated taxa or genes. This network-based approach provides deeper insights into the structure and potential roles of microbial or genes communities [17, 54]. Through WGCNA, hub ASVs—such as those annotated to Prevotella brevis, Prevotella ruminicola, Prevotella bryantii, Methanobrevibacter and Methanomassiliicoccus luminyensi were identified as key taxa driving module dynamics and contributing to CH4-related metabolic pathways. Additionally, WGCNA excels in its ability to integrate multi-effects data, enabling the associations of microbial communities with external factors such as diet, management practices, or CH4 emissions [29]. This comprehensive framework makes WGCNA a powerful tool for investigating complex relationships within microbial ecosystems and linking them to functional and environmental factors.

Additionally, our study employed both MiSeq and HiSeq sequencing platforms, two of the leading choices for short-read sequencing. These platforms are considered leading choices for various genomic and microbiome studies due to their robust data output and high-quality sequencing capabilities [55,56,57]. However, despite their widespread use, the potential influence of platform-specific differences on data analysis remains unclear in the context of our study. A study reported that HiSeq performed longer read sequences and better assigned taxa compared to MiSeq in the context of oral microbiome samples [58]. Moreover, long-read platforms such as Oxford Nanopore [59] and PacBio [60], may provide complementary strengths to short-read technologies by generating longer read lengths that can span complex genomic regions, improve genome assemblies, and resolve ambiguities in repetitive sequences [61].

While PICRUSt2 enabled functional predictions, its reliance on human-derived databases poses limitations in accurately capturing the functional potential of rumen microbes [62]. To address this, COwPi [63]—a rumen microbiome-focused adaptation of PICRUSt [64]—offers a tailored database specific to the unique microbial communities and metabolic pathways of the rumen. This targeted approach reduces mismatches and improves the accuracy of functional predictions, particularly for KEGG pathway analysis [63]. However, it is important to note that the original PICRUSt has no longer developed, limiting its applicability to current datasets and novel discoveries in rumen microbiology and impeding the usage of COwPi. Despite these challenges, PICRUSt2 remains a practical and robust tool for predicting microbial functional genes, owing to its enhanced algorithm and extended database support. Incorporating rumen-specific features from CowPi into PICRUSt2 could further refine predictions, offering a comprehensive solution for rumen microbiome research.

Despite providing valuable insights into microbial interactions related to CH4 emissions, our study has several limitations. One key limitation is our reliance on inferred functional data from 16S rRNA gene amplicon sequencing. While predictive tools such as PICRUSt2 offer valuable insights, they lack the resolution of direct metagenomic or metatranscriptomic approaches, which could provide more accurate functional annotations. Integrating multi-omics data, including metabolomics and metaproteomic, would enhance our understanding of the active metabolic pathways driving CH4 emissions.

Another limitation is the absence of longitudinal measurements. Our study provides a snapshot of microbial interactions at a single time point, but microbiota composition and CH4 emissions fluctuate over time due to factors such as diet changes, lactation stage, and seasonal variations. Longitudinal studies tracking microbial shifts across different feeding regimes and environmental conditions would be crucial to fully understanding the stability and adaptability of CH4 -related microbial networks. Additionally, while herd-specific factors were considered, more detailed environmental metadata—such as individual feeding behaviors, rumination time, and precise nutrient intake—could further clarify the external influences on microbial community structures.

Conclusions

This study employed WGCNA to investigate the interactions within the rumen microbiota and their associations with CH4 emissions using a population-level of 750 Holstein cows. The MEblue was significantly correlated with CH4 emission, revealing the critical roles of taxa such as Prevotella, Methanobrevibacter, and Methanomassiliicoccus luminyensis. These taxa underscore the complex interactions in carbohydrate fermentation and methanogenesis, key processes contributing to CH₄ production. Additionally, MEbrown was strongly associated with herd factors, revealing microbial networks influenced by farm management practices, diet composition, and feeding strategies. Functional predictions emphasized the complementary roles of bacteria and archaea in the rumen ecosystem, where bacteria provide substrates such as short-chain fatty acids (SCFAs) and hydrogen for methanogenic archaea, which are enriched in pathways linked to CH4 production. The present study highlights the microbiota-trait and microbiota-microbiota interactions related to CH4 emission in dairy cattle, contributing to a systematic understanding of CH4 production in cattle and offering key information on microbial management for mitigating environmental impact in cattle population.

Methods

Data preparation

All microbial data is same in this publication [12]. The detailed process is below. Rumen samples were drawn from individual cows using “Flora Rumen Scoop” [65], an oral insertion probe, to collect approximately 40 ml of liquid for 750 cows individually [12, 28]. To avoid cross-contamination, the “Flora Rumen Scoop” was carefully rinsed during each sampling to minimize cross-contamination. Then samples were labeled, promptly stored on ice, and transported with liquid nitrogen to the laboratory within two hours for the next steps. Each collected samples were vortexed, and a subsample of 1.2 ml liquid was contained in 1.5 ml tube and frozen in liquid nitrogen until transfer to the sequencing company (GATC, Biotech, Constance, Germany) [12]. The corresponding CH4 emission records of these 750 lactating cows are described in detail earlier [12, 28]. Fourier Transform Infrared unit (FTIR; Gasmet DX-4000, Gasmet Technologies, Helsinki, Finland) [66] and non-dispersive infrared (NDIR; Guardian NG/Gascard Edinburgh Instruments Ltd., Livingston, UK) [67] which were fitted within automatic milking machines were applied to record the CH4 emissions from the studied cows. All herds practiced indoor feeding strategies with ad libitum access to feed and water. A total mixed ration (TMR) was provided, consisting primarily of rolled barley, corn silage, grass clover silage, rapeseed meal, soybean meal, and up to 3 kg of concentrate supplement administered during milking. Although the TMR formula across all commercial herds were similar, variances in ingredients of farms were anticipated to impact the nutritional values and fiber content of the TMR across different herds.

16S rRNA gene sequencing and bioinformatics

A Qiagen QIAamp stool kit (Valencia, United States of America) was used to obtain total DNA from each rumen liquid sample [68]. Subsequently, sequencing library construction and sequencing were conducted by GATC Biotech (Constance, Germany). The variable regions V1-V3 of 16S rRNA gene for bacteria, while the variable regions V4-V6 of 16S rRNA gene for archaea were amplified using two primer sets to analyze rumen microbial profiles. Archaeal primers were S-D-Arch-0519-a-S-15: 5'-CAGCMGCCGCGGTAA-3' and S-D-Arch-1041-a-A-18 5'-GGCCATGCACCWCCTCTC-3' (expected amplicon size 542 bp), whereas bacterial primers were 27F: 5’-AGAGTTTGATCCTGGCTCAG-3’ and 534R: 5’-ATTACCGCGGCTGCTGG-3’ (expected amplicon size 508 bp) [69, 70]. To analyze regional amplicons, GoTaq Green polymerase was used for PCR amplification, and Illumina Miseq and Hiseq platforms were used to generate paired-end sequencing (2 × 300 bp). Sequencing data were processed using Usearch11 [71], Vsearch [72] and QIIME2 [73]. Initially, paired-end reads were merged by their overlapping sequences using Usearch11 (using command -fastx merge). Subsequently, primers and homopolymer sequence runs from merged sequences were trimmed by Usearch11 -search oligodb algorithm. Quality control was executed by Vsearch –fastq eestats2 and Usearch11 -fastx truncate. Only sequences ≥ 300 bp and ≥ 450 bp in length for archaea and bacteria, respectively, were retained for subsequent analyses. The identification of amplicon sequence variants (ASVs) was conducted using DADA2 [74] in QIIME2. The ASVs table for archaea and bacteria were filtered by QIIME2 using feature-table command line with filter-samples to eliminate low counts (> 2,000 reads for bacteria and > 1,000 reads for archaea) and using filter-features to trim low relative abundance (> 0.0001 for each). Subsequently, QIIME diversity subsample-table was used to normalize archaea and bacteria ASV tables. Then the QIIME diversity alpha was conducted to estimate the α-diversity indices (Chao1, Shannon 2, Simpson, richness, and reads). β-diversity (Principal Component Analysis [PCA]) was calculated using Bray–Curtis dissimilarity matrices using phyloseq package in R [75], confirmed with PERMANOVA. Moreover, the representative sequences for archaea and bacteria ASVs were annotated at domain, phylum, class, order, family, genus, and species levels to the database Greengenes2 [76] using QIIME greengenes2 taxonomy-from-table. We then predicted functional characteristics of general archaea and bacteria communities, WGCNA modules, and their hub ASVs using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt2 V2.2.0) [62] with Kyoto Encyclopedia of Genes and Genomes (KEGG) [77] database. It predicted the microbial function according to the proportion of marker gene sequences in samples [62]. Only taxa with a relative abundance > 0.01% among half samples were considered as detected taxa and included in the downstream analyses [12]. Subsequently, ASVs units were visualized using phylogenetic trees generated through the application of the Metacoder package in the R software [78]. The relative abundance of taxonomic annotation and KEGG pathways for archaea and bacteria were plotted by ggplot2 in R [79]. After quality trimming, the ASVs of archaea and bacteria were disseminated by principle component analysis (PCA) using Vegan [26] package in R, considering the influence of herds and lactations.

CH4 records and correction

The mean of 750 Holstein CH4 concentrations were corrected for environmental factors, including diurnal variation and day to day differences using a linear mixed model following a previous study [67]. Then the CH4 emission records were corrected by a linear mixed model including fixed effects of herd (Herd), lactation (Lact), and days in milk (DIM) using R package lme4 [80]. The models are described below.

1) Fixed effect model for testing effects of herd (Herd), lactation (Lact), and days in milk (DIM) on CH4 emission:

$${y}_{ijk}=\mu +{Herd}_{i}+{Lact}_{j}+\beta .{DIM}_{ijk}+{\varepsilon }_{ijk}$$

\({y}_{ijk}\): Observed CH4 emissions for the k-th individual in the i-th Herd and j-th Lactation.

μ: Overall mean.

\({Herd}_{i}\): Fixed effect of the i-th Herd (i = 1,2,…,6).

\({Lact}_{j}\): Fixed effect of the j-th Lactation (j = 1,2).

\(\beta\): Regression coefficient for Days in Milk (DIM).

\({DIM}_{ijk}\): Days in Milk for the k-th individual in the i-th Herd and j-th Lactation.

\({\varepsilon }_{ijk}\): Residual error term, assumed to follow \(N(0, {\sigma }_{e}^{2})\), \({\sigma }_{e}^{2}\) is the random error variance.

2) Mixed model to get correct phenotype for CH4

$${y}_{ijk}=\mu +{Herd}_{i}+{Lact}_{j}+\beta .{DIM}_{ijk}+{u}_{k}+{\varepsilon }_{ijk}$$

\({y}_{ijk}\): Corrected CH4 emissions for the k-th individual in the i-th Herd and j-th Lactation.

μ: Overall mean.

\({Herd}_{i}\): Fixed effect of the i-th Herd (i = 1,2,…,6).

\({Lact}_{j}\): Fixed effect of the j-th Lactation (j = 1,2).

\(\beta\): Regression coefficient for Days in Milk (DIM).

\({DIM}_{ijk}\): Days in Milk for the k-th individual in the i-th Herd and j-th Lactation.

\({u}_{k}\): Random effect of the k-th individual, assumed to follow\(N(0, {\sigma }_{u}^{2})\), \({\sigma }_{u}^{2}\) is the random effect of the cows.

\({\varepsilon }_{ijk}\): Residual error term, assumed to follow \(N(0, {\sigma }_{e}^{2})\), \({\sigma }_{e}^{2}\) is the random error variance.

Co-expression network of rumen microbiota and traits association.

Co-expression network among detected bacterial and archaeal at genus-level taxa were inferred using the Weighted Gene Co-expression Network Analysis (WGCNA) implemented in R [29]. Before starting analyses, we merged archaea and bacteria ASVs. We then used WGCNA to identify the modules of ASVs with the highest relevance to lactations (Lact), Herd, DIM, and CH4 emissions from cows. The normalized rumen microbial data were log 10 transferred and the combined dataset was quality-checked before analyzing. The low-quality samples and ASVs were removed from the combined dataset using function goodSamplesGenes(). A threshold power of 10 for microbial ASVs data were chosen in the quality control considering that they were the smallest threshold that results in a scale-free R2 fit of 0.8. After quality control, WGCNA was applied to the combined dataset to generate the network, using blockwiseModule() function. The blockwiseModule() function was used to calculate topologic overlap matrix (TOM) with correlation function, followed by ASVs being hierarchically clustered using 1-TOM (dissTOM) as the distance measure. Original module assignments were determined by using dynamaic tree-cutting, using default parameters mergeCutHeight = 0.25, and minModulesize = 30. The clustered modules were plotted by function plotDendroAndColors() with clustering dendrogram and module colors. The Lact, Herd, DIM, and CH4 data were used to select the highly correlated modules. The moduleEigengenes() function was used to calculate the module eigengenes (MEs) for each ASV. We then calculated the correlations of MEs with Lact, Herd, DIM, and CH4 using corPvalueStudent() function. The hub ASVs were selected by function intramodularConnectivity() using dissTOM and moduleColors as input. All hub ASVs interaction were then displayed by networks using Cytoscape [81]. Subsequently, the significant modules ASV sequences and their hub ASV sequences combined with their feature tables were inputted to PICRUSt2 to predict the microbial functions based on KEGG database, as in the previous step. The relative abundance of KEGG level-3 pathways were plotted by ggplot2 package in R [79].

Availability of data and materials

All microbiota sequence data is freely available at https://www.ebi.ac.uk/ena/data/view/PRJEB28065. The data underlying this article are available in G. F. Difford and Q. Zhang, at https://doiorg.publicaciones.saludcastillayleon.es/https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pgen.1007580 and https://doiorg.publicaciones.saludcastillayleon.es/https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41396-020-0663-x.

Abbreviations

ASV:

Amplicon sequence variant

CH4 :

Methane

DIM:

Days in milk

GHG:

Greenhouse gas

Lact:

Lactation numbers of each cow

MEs:

Module eigengenes

PCA:

Principal component analysis

TOM:

Topologic overlap matrix

WGCNA:

Weighted gene co-expression network analysis

References

  1. Filonchyk M, Peterson MP, Zhang L, Hurynovich V, He Y. Greenhouse gases emissions and global climate change: examining the influence of CO2, CH4, and N2O. Sci Total Environ. 2024;935: 173359.

    Article  CAS  PubMed  Google Scholar 

  2. OECD/FAO: OECD-FAO Agricultural Outlook 2023–2032. (Publishing O ed. Paris; 2023.

  3. Gerber PJ, Steinfeld H, Henderson B, Mottet A, Opio C, Dijkman J, Falcucci A, Tempio G: Tackling climate change through livestock: a global assessment of emissions and mitigation opportunities. Food and agriculture Organization of the United Nations (FAO); 2013.

  4. FAO, GDP: Climate change and the global dairy cattle sector-The role of the dairy sector in a low-carbon future., vol. 36 pp. Rome; 2018.

  5. FAO: Pathways towards lower emissions – A global assessment of the greenhouse gas emissions and mitigation options from livestock agrifood systems. Rome; 2023.

  6. Brede J, Peukert M, Egert B, Breves G, Brede M. Long-term mootral application impacts methane production and the microbial community in the rumen simulation technique system. Front Microbiol. 2021;12:691502.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Fresco S, Boichard D, Fritz S, Lefebvre R, Barbey S, Gaborit M, Martin P. Comparison of methane production, intensity, and yield throughout lactation in Holstein cows. J Dairy Sci. 2023;106:4147–57.

    Article  CAS  PubMed  Google Scholar 

  8. Hristov AN, Melgar A, Wasson D, Arndt C. Symposium review: Effective nutritional strategies to mitigate enteric methane in dairy cattle. J Dairy Sci. 2022;105:8543–57.

    Article  CAS  PubMed  Google Scholar 

  9. Worku D. Unraveling the genetic basis of methane emission in dairy cattle: a comprehensive exploration and breeding approach to lower methane emissions. Anim Biotechnol. 2024;35:2362677.

    Article  PubMed  Google Scholar 

  10. Matthews C, Crispie F, Lewis E, Reid M, O’Toole PW, Cotter PD. The rumen microbiome: a crucial consideration when optimising milk and meat production and nitrogen utilisation efficiency. Gut Microbes. 2019;10:115–32.

    Article  CAS  PubMed  Google Scholar 

  11. Tapio I, Snelling TJ, Strozzi F, Wallace RJ. The ruminal microbiome associated with methane emissions from ruminant livestock. J Anim Sci Biotechnol. 2017;8:7.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Difford GF, Plichta DR, Løvendahl P, Lassen J, Noel SJ, Højberg O, Wright A-D, Zhu Z, Kristensen L, Nielsen HB, Guldbrandtsen B, Sahana G. Host genetics and the rumen microbiome jointly associate with methane emissions in dairy cows. PLoS Genet. 2018;14(10):e1007580. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pgen.1007580.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Gillah KA, Kifaro GC, Madsen J. Effects of management practices on yield and quality of milk from smallholder dairy units in urban and peri-urban Morogoro. Tanzania Tropical Animal Health Production. 2014;46:1177–83.

    Article  PubMed  Google Scholar 

  14. Sova AD, LeBlanc SJ, McBride BW, DeVries TJ. Associations between herd-level feeding management practices, feed sorting, and milk production in freestall dairy farms. J Dairy Sci. 2013;96:4759–70.

    Article  CAS  PubMed  Google Scholar 

  15. van Breukelen AE, Veerkamp RF, de Haas Y, Aldridge MN. Genetic parameter estimates for methane emission from breath during lactation and potential inaccuracies in reliabilities assuming a repeatability versus random regression model. J Dairy Sci. 2024;107:5853–68.

    Article  PubMed  Google Scholar 

  16. Henderson G, Cox F, Ganesh S, Jonker A, Young W, Abecia L, Angarita E, Aravena P, Nora Arenas G, Ariza C, et al. Rumen microbial community composition varies with diet and host, but a core microbiome is found across a wide geographical range. Sci Rep. 2015;5:14567.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Martínez-Álvaro M, Auffret MD, Duthie C-A, Dewhurst RJ, Cleveland MA, Watson M, Roehe R. Bovine host genome acts on rumen microbiome function linked to methane emissions. Commun Biology. 2022;5:350.

    Article  Google Scholar 

  18. Stewart RD, Auffret MD, Warr A, Wiser AH, Press MO, Langford KW, Liachko I, Snelling TJ, Dewhurst RJ, Walker AW. Assembly of 913 microbial genomes from metagenomic sequencing of the cow rumen. Nat Commun. 2018;9:870.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Martínez-Álvaro M, Auffret MD, Stewart RD, Dewhurst RJ, Duthie CA, Rooke JA, Wallace RJ, Shih B, Freeman TC, Watson M, Roehe R. Identification of complex rumen microbiome interaction within diverse functional niches as mechanisms affecting the variation of methane emissions in Bovine. Front Microbiol. 2020;11:659.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Khairunisa BH, Heryakusuma C, Ike K, Mukhopadhyay B, Susanti D. Evolving understanding of rumen methanogen ecophysiology. Front Microbiol. 2023;6(14):1296008.

    Article  Google Scholar 

  21. Toyber I, Kumar R, Jami E. Rumen protozoa are a hub for diverse hydrogenotrophic functions. Environ Microbiol Rep. 2024;16: e13298.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Black JL, Davison TM, Box I. Methane emissions from ruminants in Australia: mitigation potential and applicability of mitigation strategies. Animals. 2021;11(4):951.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Islam M, Lee SS. Advanced estimation and mitigation strategies: a cumulative approach to enteric methane abatement from ruminants. J Anim Sci Technol. 2019;61:122–37.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Smith PE, Kelly AK, Kenny DA, Waters SM. Enteric methane research and mitigation strategies for pastoral-based beef cattle production systems. Front Veterinary Sci. 2022;23(9):958340.

    Article  Google Scholar 

  25. Wolff SM, Ellison MJ, Hao Y, Cockrum RR, Austin KJ, Baraboo M, Burch K, Lee HJ, Maurer T, Patil R, et al. Diet shifts provoke complex and variable changes in the metabolic networks of the ruminal microbiome. Microbiome. 2017;5:60.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Dixon P. VEGAN, a package of R functions for community ecology. J Veg Sci. 2003;14:927–30.

    Article  Google Scholar 

  27. Roehe R, Dewhurst RJ, Duthie C-A, Rooke JA, McKain N, Ross DW, Hyslop JJ, Waterhouse A, Freeman TC, Watson M, Wallace RJ. Bovine Host Genetic Variation Influences Rumen Microbial Methane Production with best selection criterion for low methane emitting and efficiently feed converting hosts based on metagenomic gene abundance. PLoS Genet. 2016;12: e1005846.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Zhang Q, Difford G, Sahana G, Løvendahl P, Lassen J, Lund MS, Guldbrandtsen B, Janss L. Bayesian modeling reveals host genetics associated with rumen microbiota jointly influence methane emission in dairy cows. ISME J. 2020;14:2019–33.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9:559.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Zhang FL, Li WD, Zhang G, Zhang M, Liu ZJ, Zhu KX, Liu QC, Zhang SE, Shen W, Zhang XF. Identification of unique transcriptomic signatures through integrated multispecies comparative analysis and WGCNA in bovine oocyte development. BMC Genomics. 2023;24:1–12.

    Google Scholar 

  31. Guo X, Zhang H, Wang H, He X-X, Wang J-X, Wei W, Liu M, Xu J-M, Liu Y-N, Jiang R-S. Identification of key modules and hub genes involved in regulating the color of chicken breast meat using WGCNA. Animals. 2023;13:2356.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Volmer JG, McRae H, Morrison M. The evolving role of methanogenic archaea in mammalian microbiomes. Front Microbiol. 2023;14:1268451.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Patra A, Park T, Kim M, Yu Z. Rumen methanogens and mitigation of methane emission by anti-methanogenic compounds and substances. J Animal Sci Biotechnol. 2017;8:13.

    Article  Google Scholar 

  34. Wallace RJ, Rooke JA, McKain N, Duthie C-A, Hyslop JJ, Ross DW, Waterhouse A, Watson M, Roehe R. The rumen microbial metagenome associated with high methane production in cattle. BMC Genomics. 2015;16:839.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Feldewert C, Lang K, Brune A. The hydrogen threshold of obligately methyl-reducing methanogens. FEMS Microbiol Lett. 2020;367:fnaa137.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Xie F, Zhao S, Zhan X, Zhou Y, Li Y, Zhu W, Pope PB, Attwood GT, Jin W, Mao S. Unraveling the phylogenomic diversity of Methanomassiliicoccales and implications for mitigating ruminant methane emissions. Genome Biol. 2024;25:32.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Fregulia P, Dias RJP, Campos MM, Tomich TR, Pereira LGR, Neves ALA. Composition of the rumen microbiome and its association with methane yield in dairy cattle raised in tropical conditions. Mol Biol Rep. 2024;51:447.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Aguilar-Marin SB, Betancur-Murillo CL, Isaza GA, Mesa H, Jovel J. Lower methane emissions were associated with higher abundance of ruminal Prevotella in a cohort of Colombian buffalos. BMC Microbiol. 2020;20:1–13.

    Article  Google Scholar 

  39. Betancur-Murillo CL, Aguilar-Marín SB, Jovel J: Prevotella: A Key Player in Ruminal Metabolism. In Microorganisms, vol. 11; 2023.

  40. Zhang L, Ren W, Bi Y, Zhang J, Cheng Y, Xu X. Effects of different feeding patterns on the rumen bacterial community of tan lambs, based on high-throughput sequencing of 16S rRNA amplicons. Front Microbiol. 2023;14:1228935.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Yang G, Hong S, Yang P, Sun Y, Wang Y, Zhang P, Jiang W, Gu Y. Discovery of an ene-reductase for initiating flavone and flavonol catabolism in gut bacteria. Nat Commun. 2021;12:790.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Palmonari A, Federiconi A, Formigoni A. Animal board invited review: The effect of diet on rumen microbial composition in dairy cows. Animal. 2024;18:101319.

    Article  CAS  PubMed  Google Scholar 

  43. Malik PK, Trivedi S, Mohapatra A, Kolte AP, Sejian V, Bhatta R, Rahman H. Comparison of enteric methane yield and diversity of ruminal methanogens in cattle and buffaloes fed on the same diet. PLoS ONE. 2021;16: e0256048.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Kim S-H, Mamuad LL, Islam M, Lee S-S. Reductive acetogens isolated from ruminants and their effect on <italic>in vitro</italic> methane mitigation and milk performance in Holstein cows. J Animal Sci Technol. 2020;62:1–13.

    Article  CAS  Google Scholar 

  45. Vargas JE, Andrés S, López-Ferreras L, Snelling TJ, Yáñez-Ruíz DR, García-Estrada C, López S. Dietary supplemental plant oils reduce methanogenesis from anaerobic microbial fermentation in the rumen. Sci Rep. 2020;10:1613.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. van Zijderveld SM, Gerrits WJ, Apajalahti JA, Newbold JR, Dijkstra J, Leng RA, Perdok HB. Nitrate and sulfate: Effective alternative hydrogen sinks for mitigation of ruminal methane production in sheep. J Dairy Sci. 2010;93:5856–66.

    Article  PubMed  Google Scholar 

  47. Patra AK, Saxena J. The effect and mode of action of saponins on the microbial populations and fermentation in the rumen and ruminant production. Nutr Res Rev. 2009;22:204–19.

    Article  CAS  PubMed  Google Scholar 

  48. Tavendale MH, Meagher LP, Pacheco D, Walker N, Attwood GT, Sivakumaran S. Methane production from in vitro rumen incubations with Lotus pedunculatus and Medicago sativa, and effects of extractable condensed tannin fractions on methanogenesis. Anim Feed Sci Technol. 2005;123:403–19.

    Article  Google Scholar 

  49. Nedelkov K, Angelova T, Krastanov J, Mihaylova M. Feeding strategies to reduce methane emissions: A review. Bulgarian J Agr Sci. 2024;30:28–36.

    Google Scholar 

  50. Taxis TM, Wolff S, Gregg SJ, Minton NO, Zhang C, Dai J, Schnabel RD, Taylor JF, Kerley MS, Pires JC. The players may change but the game remains: network analyses of ruminal microbiomes suggest taxonomic differences mask functional similarity. Nucleic Acids Res. 2015;43:9600–12.

    CAS  PubMed  PubMed Central  Google Scholar 

  51. Li F, Guan LL. Metatranscriptomic profiling reveals linkages between the active rumen microbiome and feed efficiency in beef cattle. Appl Environ Microbiol. 2017;83:e00061-e117.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Puente-Sánchez F, Pascual-García A, Bastolla U, Pedrós-Alió C, Tamames J. Cross-biome microbial networks reveal functional redundancy and suggest genome reduction through functional complementarity. Commun Biol. 2024;7:1046.

    Article  PubMed  PubMed Central  Google Scholar 

  53. Friedman J, Alm EJ. Inferring correlation networks from genomic survey data. PLoS Comput Biol. 2012;8: e1002687.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Li Y, Mao K, Zang Y, Lu G, Qiu Q, Ouyang K, Zhao X, Song X, Xu L, Liang H, Qu M. Revealing the developmental characterization of rumen microbiome and its host in newly received cattle during receiving period contributes to formulating precise nutritional strategies. Microbiome. 2023;11:238.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  55. Salamon D, Zapała B, Krawczyk A, Potasiewicz A, Nikiforuk A, Stój A, Gosiewski T. Comparison of iSeq and MiSeq as the two platforms for 16S rRNA sequencing in the study of the gut of rat microbiome. Appl Microbiol Biotechnol. 2022;106:7671–81.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Hu T, Chen J, Lin X, He W, Liang H, Wang M, Li W, Wu Z, Han M, Jin X, et al. Comparison of the DNBSEQ platform and Illumina HiSeq 2000 for bacterial genome assembly. Sci Rep. 2024;14:1292.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Castelino M, Eyre S, Moat J, Fox G, Martin P, Ho P, Upton M, Barton A. Optimisation of methods for bacterial skin microbiome investigation: primer selection and comparison of the 454 versus MiSeq platform. BMC Microbiol. 2017;17:23.

    Article  PubMed  PubMed Central  Google Scholar 

  58. Hee Sam N, et al. Comparison of the performance of MiSeq and HiSeq 2500 in a microbiome study. Microbiol Biotechnol Lett. 2020;48:574–81.

    Article  Google Scholar 

  59. Wang Y, Zhao Y, Bollas A, Wang Y, Au KF. Nanopore sequencing technology, bioinformatics and applications. Nat Biotechnol. 2021;39:1348–65.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Rhoads A, Au KF. PacBio sequencing and its applications. Genom Proteom Bioinform. 2015;13:278–89.

    Article  Google Scholar 

  61. Logsdon GA, Vollger MR, Eichler EE. Long-read human genome sequencing and its applications. Nat Rev Genet. 2020;21:597–614.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  62. Douglas GM, Maffei VJ, Zaneveld JR, Yurgel SN, Brown JR, Taylor CM, Huttenhower C, Langille MGI. PICRUSt2 for prediction of metagenome functions. Nat Biotechnol. 2020;38:685–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Wilkinson TJ, Huws SA, Edwards JE, Kingston-Smith AH, Siu-Ting K, Hughes M, Rubino F, Friedersdorff M, Creevey CJ. CowPI: A rumen microbiome focussed version of the PICRUSt functional inference software. Front Microbiol. 2018;9:1095.

    Article  PubMed  PubMed Central  Google Scholar 

  64. Langille MGI, Zaneveld J, Caporaso JG, McDonald D, Knights D, Reyes JA, Clemente JC, Burkepile DE, Vega Thurber RL, Knight R, et al. Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences. Nat Biotechnol. 2013;31:814–21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  65. Geishauser T, Linhart N, Neidl A, Reimann A. Factors associated with ruminal pH at herd level. J Dairy Sci. 2012;95:4556–67.

    Article  CAS  PubMed  Google Scholar 

  66. Lassen J, Løvendahl P. Heritability estimates for enteric methane emissions from Holstein cattle measured using noninvasive methods. J Dairy Sci. 2016;99:1959–67.

    Article  CAS  PubMed  Google Scholar 

  67. Difford GF, Lassen J, Løvendahl P. Interchangeability between methane measurements in dairy cows assessed by comparing precision and agreement of two non-invasive infrared methods. Comput Electron Agric. 2016;124:220–6.

    Article  Google Scholar 

  68. Johnson DJ, Martin LR, Roberts KA. STR-typing of human DNA from human fecal matter using the QIAGEN QIAamp stool mini kit. J Forensic Sci. 2005;50:802–8.

    Article  CAS  PubMed  Google Scholar 

  69. Balasubramanian D, Schneper L, Kumari H, Mathee K. A dynamic and intricate regulatory network determines Pseudomonas aeruginosa virulence. Nucleic Acids Res. 2013;41:1–20.

    Article  CAS  PubMed  Google Scholar 

  70. Jumpstart Consortium Human Microbiome Project Data Generation Working G: Evaluation of 16S rDNA-Based Community Profiling for Human Microbiome Research. PLOS ONE 2012, 7:39315.

  71. Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 2010;26:2460–1.

    Article  CAS  PubMed  Google Scholar 

  72. Rognes T, Flouri T, Nichols B, Quince C, Mahé F. VSEARCH: a versatile open source tool for metagenomics. PeerJ. 2016;4: e2584.

    Article  PubMed  PubMed Central  Google Scholar 

  73. Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet CC, Al-Ghalith GA, Alexander H, Alm EJ, Arumugam M, Asnicar F, et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol. 2019;37:852–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  74. Robert CE: UNOISE2: improved error-correction for Illumina 16S and ITS amplicon sequencing. bioRxiv 2016:081257.

  75. McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE. 2013;8: e61217.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  76. McDonald D, Jiang Y, Balaban M, Cantrell K, Zhu Q, Gonzalez A, Morton JT, Nicolaou G, Parks DH, Karst SM, et al. Greengenes2 unifies microbial data in a single reference tree. Nat Biotechnol. 2024;42:715–8.

    Article  CAS  PubMed  Google Scholar 

  77. Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28:27–30.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  78. Foster ZSL, Sharpton TJ, Grünwald NJ. Metacoder: An R package for visualization and manipulation of community taxonomic diversity data. PLoS Comput Biol. 2017;13: e1005404.

    Article  PubMed  PubMed Central  Google Scholar 

  79. Wickham H. ggplot2: Elegant Graphics for Data Analysis. New York: Springer-Verlag; 2016.

    Book  Google Scholar 

  80. Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. J Stat Softw. 2015;67:1–48.

    Article  Google Scholar 

  81. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13:2498–504.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank for the project REMRUM (Project No. 11-105913) funded by the Danish Strategic Research Council, Danish Research Council for Independent Research, Technology and Production. Bingjie Li acknowledges funding from the UK Biotechnology and Biological Sciences Research Council (BBSRC) grant BB/X009505/1. Meanwhile, we are grateful to our reviewers and editor for the helpful comments to improve the manuscript.

Funding

This study has no funding to support.

Author information

Author notes

  1. Bingjie Li and Zexi Cai have contributed equally to this work.

Authors and Affiliations

  1. Center for Quantitative Genetics and Genomics, Aarhus University, CF Møllers Allé 3, 8000, Aarhus, Denmark

    Xiaoxing Ye, Goutam Sahana, Mogens Sandø Lund & Zexi Cai

  2. Department of Animal and Veterinary Sciences, Scotland’s Rural College (SRUC), Edinburgh, UK

    Bingjie Li

Authors
  1. Xiaoxing Ye
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Goutam Sahana
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Mogens Sandø Lund
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Bingjie Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Zexi Cai
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

XY analyzed and interpreted data. XY wrote the manuscript. BL, ZC modified analyzing process and manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Xiaoxing Ye.

Ethics declarations

Ethics approval and consent to participate

All animals handlings were conducted according to “Metagenomics in Dairy Cows” protocol. The protocol was approved by The Animal Experiments Inspectorate, Danish Veterinary and Food Administration, Ministry of Environment and Food of Denmark (Approval number 2016–15-0201–00959).

Consent for publication

The dataset used in this study was generated previously. No animal handling was involved in this study.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ye, X., Sahana, G., Lund, M.S. et al. Network analyses unraveled the complex interactions in the rumen microbiota associated with methane emission in dairy cattle. anim microbiome 7, 24 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s42523-025-00386-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s42523-025-00386-z

Keywords

Animal Microbiome

ISSN: 2524-4671