Metagenomics vs Genomics: Key Differences and Complementary Roles

Genomics and metagenomics are distinct sequencing-based approaches for studying genetic material. Genomics examines the complete DNA content of a single organism, typically an isolate or a host, to understand its genes, mutations, and biological functions. Metagenomics analyzes the collective genetic material recovered directly from an environmental or clinical sample, such as soil, water, feces, or a wound swab, without the need to culture individual organisms. The practical question for researchers and analysts is not which method is superior, but which approach answers the biological question at hand, and how the two can be combined to provide a more complete picture. This article compares the conceptual foundations, methodological workflows, data requirements, and application contexts of genomics and metagenomics, and provides concrete decision criteria for selecting and integrating these approaches.

Scope and Definitions

Genomics focuses on the structure, function, and evolution of a single genome. The starting material is usually a pure culture, a single tissue sample, or a host organism. The goal is to produce a high-quality genome assembly that represents one species or one individual. Applications include identifying disease-causing mutations in a patient, characterizing a bacterial isolate for antimicrobial resistance, or assembling the genome of a newly discovered eukaryote.

Metagenomics starts with a mixed community. The sample contains DNA from many organisms, including bacteria, archaea, viruses, fungi, and sometimes the host itself. The goal is to describe the composition, diversity, and functional potential of that community. Metagenomics can be used to detect pathogens in clinical specimens, profile the gut microbiome in health and disease, or monitor microbial communities in environmental and agricultural systems.

The distinction matters for every downstream decision, from DNA extraction to sequencing depth to bioinformatics analysis. A genomics workflow assumes a single dominant genome and can tolerate high sequencing depth per base. A metagenomics workflow must handle uneven abundance, contamination from host DNA, and the absence of a single reference genome.

Core Principles of Genomics

Genomics is built on the assumption that the sample contains one genome of interest. The workflow begins with DNA extraction optimized for high yield and high molecular weight, followed by library preparation and sequencing. Short-read platforms such as Illumina produce high accuracy but fragmented assemblies. Long-read platforms such as Oxford Nanopore Technologies and PacBio produce longer contiguous sequences that resolve repetitive regions and structural variants.

The analysis pipeline for a single genome includes quality trimming, assembly, scaffolding, gene prediction, and functional annotation. For bacterial isolates, additional steps include multilocus sequence typing, antimicrobial resistance gene detection, and phylogenetic placement. For human genomes, the pipeline includes variant calling against a reference genome and interpretation of clinical significance.

A key strength of genomics is the ability to produce a complete or near-complete reference genome. This enables precise identification of genes, operons, regulatory elements, and mobile genetic elements. It also supports comparative genomics, where multiple isolates of the same species are compared to track transmission, evolution, or resistance spread. For example, population genomics of Escherichia coli in wastewater and river environments can reveal how antimicrobial resistance dynamics change across environmental compartments [27].

Genomics has limitations. It requires culturing or physical isolation of the organism, which is impossible for many environmental microbes. It also provides no information about the community context, such as which other species are present or how the target organism interacts with them.

Core Principles of Metagenomics

Metagenomics does not require culture or isolation. DNA is extracted directly from the sample, and all genetic material is sequenced together. The resulting data set contains fragments from many genomes at varying abundances. The analysis can be read-based or assembly-based.

Read-based metagenomics assigns individual sequencing reads to taxonomic groups by comparing them against reference databases. This approach is fast and works well when the community is dominated by known species. It is limited by the completeness of the reference database and cannot resolve strain-level variation or discover novel species.

Assembly-based metagenomics reconstructs genomes from the mixed data set. Reads are assembled into contigs, and contigs are grouped into metagenome-assembled genomes (MAGs) based on coverage and composition. MAGs can represent previously uncharacterized species and enable functional analysis of unculturable organisms. The quality of MAGs varies, and standards for completion and contamination are essential for reliable interpretation.

A scalable genome-resolved framework applied to 1,878 deeply sequenced samples from the Estonian Microbiome Cohort reconstructed 84,762 MAGs representing 2,257 species, including 353 previously uncharacterized species that reached up to 30% relative abundance in some individuals [16]. This demonstrates that metagenomics can discover novel biology that genomics alone cannot access.

Metagenomics also enables functional profiling. By annotating genes in the community, researchers can assess the metabolic potential of the microbiome, including pathways for carbohydrate degradation, vitamin synthesis, and antimicrobial resistance. Shotgun metagenomic profiling of respiratory and blood samples from Gambian children with pneumonia recovered 60 medium-quality and 35 high-quality MAGs, including 11 Streptococcus pneumoniae and 10 Haemophilus influenzae strains, and showed that more than 70% of detected antimicrobial resistance genes were found exclusively in a single species [17].

At a Glance

The table below summarizes the key differences between genomics and metagenomics across scope, methodology, and application.

Methodological Workflow Comparison

Sample Collection and Storage

Genomics samples are collected with the expectation of isolating a single organism. For bacterial isolates, a pure culture is grown and harvested. For host genomics, tissue or blood is collected and processed to enrich for host cells. Sample storage conditions depend on the downstream application, but DNA stability is the primary concern.

Metagenomics samples require careful handling to preserve the full community composition. Different organisms lyse at different rates, and storage conditions can shift the community profile. For example, fecal samples are often frozen immediately to minimize bacterial growth and DNA degradation. The choice of preservation buffer and storage temperature should be validated for the sample type.

DNA Extraction

DNA extraction is a critical divergence point. For genomics, the goal is high yield and high molecular weight DNA from a single source. For metagenomics, the goal is unbiased lysis across all cell types, including gram-positive bacteria, fungi, and viruses, which have different cell wall structures.

A comparison of six DNA extraction methods for shotgun metagenomics Nanopore sequencing found that the Quick-DNA HMW MagBead Kit produced the best yield of pure high molecular weight DNA and enabled accurate detection of almost all bacterial species in a complex mock community [18]. This highlights that extraction method choice directly affects sequencing outcomes and downstream interpretation.

For long-read metagenomics, high molecular weight DNA is essential. Fragmented DNA reduces read length and compromises assembly quality. Researchers should assess DNA quantity, purity, and fragment size before library preparation.

Sequencing Platform Selection

Genomics can be performed on short-read or long-read platforms. Short reads are cost-effective and accurate but produce fragmented assemblies. Long reads resolve repeats and structural variants but have higher error rates and cost. Many genomics projects use a hybrid approach, combining short reads for accuracy and long reads for contiguity.

Metagenomics has additional constraints. The sample contains many genomes at different abundances, so sequencing depth must be sufficient to detect low-abundance organisms. Long-read metagenome assembly has advanced significantly. The myloasm assembler, designed for modern long reads such as PacBio HiFi and Oxford Nanopore Technologies R10.4, assembled three times more complete circular contigs than the next-best assembler on real-world ONT metagenomes [15]. It also recovered six complete Prevotella copri single-contig genomes from a gut metagenome and eight complete TM7 contigs with more than 93% similarity from an oral metagenome [15].

Bioinformatics Pipelines

Genomics pipelines are mature and standardized. Tools for quality control, assembly, annotation, and variant calling are well documented. The main challenges are genome size, ploidy, and repetitive content.

Metagenomics pipelines are more complex and less standardized. The choice of pipeline affects results, and reproducibility requires careful documentation of parameters and versions. The TOFU-MAaPO workflow provides a portable, automated single-command Nextflow pipeline for large-scale metagenomic analysis, and it yielded 12% to 77% more high-quality MAGs than three established pipelines in a benchmark [13]. It also automatically downloaded and taxonomically annotated 16,462 human gut metagenome samples from the Sequence Read Archive in less than 55 hours [13].

Applications of Genomics

Clinical Diagnostics and Precision Medicine

Genomics is used to identify disease-causing mutations in patients with suspected genetic disorders. Whole-exome sequencing and whole-genome sequencing can detect single nucleotide variants, insertions and deletions, and structural variants. The clinical utility depends on the availability of reference genomes and the interpretation of variants of unknown significance.

Genomics also supports pharmacogenomics, where genetic variants influence drug metabolism and response. This enables personalized treatment decisions based on the patient's genetic profile.

Pathogen Characterization and Outbreak Investigation

For infectious disease outbreaks, genomics of isolated pathogens provides high-resolution typing. Whole-genome sequencing of bacterial isolates can distinguish closely related strains, track transmission chains, and identify antimicrobial resistance determinants. Population genomics of Escherichia coli in wastewater and river environments has been used to monitor antimicrobial resistance dynamics across environmental compartments [27].

The limitation is that genomics requires culture or isolation. For pathogens that are difficult to culture, or when the patient has already received antibiotics, genomics may fail to identify the causative agent.

Eukaryotic Genome Assembly

Genomics is the standard approach for assembling the genomes of plants, animals, fungi, and protists. The quality of the assembly depends on genome size, heterozygosity, and repetitive content. Long-read sequencing has enabled chromosome-level assemblies for many species.

However, some eukaryotes cannot be purified from their hosts or environments. In these cases, metagenomics can be used to assemble a genome from an unpurified sample. Metagenomic techniques enabled the assembly of the genome of Polymyxa betae, a plasmodiophorid parasite, from unpurified zoospore holobiont, including the first mitochondrial genome for this species [24]. This demonstrates that the boundary between genomics and metagenomics is not absolute.

Applications of Metagenomics

Clinical Pathogen Detection

Metagenomic next-generation sequencing (mNGS) is widely used to detect pathogens in clinical specimens, particularly when conventional tests are negative or when the patient is critically ill. Shotgun metagenomic sequencing of bronchoalveolar lavage fluid can identify bacteria, viruses, fungi, and parasites in a single assay.

A multicenter randomized controlled trial in 10 ICUs compared mNGS combined with conventional microbiological tests to conventional tests alone in 349 patients with severe community-acquired pneumonia. The time to clinical improvement was better in the mNGS group, with a median of 10 days versus 13 days, and the proportion of patients with clinical improvement within 14 days was significantly higher in the mNGS group at 62.0% versus 46.5% [8].

The diagnostic performance of mNGS depends on the specimen type and clinical context. In a study of 511 clinical specimens, mNGS had a sensitivity of 50.7% and specificity of 85.7% for diagnosing infectious disease, outperforming culture for Mycobacterium tuberculosis, viruses, anaerobes, and fungi [12]. Importantly, mNGS was less affected by prior antibiotic exposure, with a sensitivity of 52.5% versus 34.2% for culture in cases with antibiotic exposure [12].

Targeted Next-Generation Sequencing

Because mNGS is complex and expensive, targeted next-generation sequencing (tNGS) has emerged as an alternative. tNGS uses multiplex PCR or hybrid capture to enrich for specific pathogen panels before sequencing. A study comparing tNGS to mNGS for lower respiratory tract infections found that multiplex PCR-based tNGS and hybrid capture-based tNGS took 10.3 and 16 hours, respectively, with sequencing data sizes of 0.1 million and 1 million reads, and test costs reduced to a quarter and half of mNGS [5]. The sensitivities were 86.5% and 87.3% versus 85.5% for mNGS, and specificities were 90.0% and 88.0% versus 92.1% [5].

The tradeoff is that tNGS only detects organisms in the targeted panel. In the same study, mNGS detected six samples with filamentous fungi that were missed by tNGS, while tNGS detected Pneumocystis jirovecii in seven samples that mNGS missed [5]. The choice between mNGS and tNGS depends on the clinical question, the urgency of the result, and the available budget.

Microbiome Research

Metagenomics is the primary tool for studying the human microbiome. Shotgun metagenomic sequencing provides species-level resolution and functional information that 16S rRNA gene sequencing cannot achieve. A meta-analysis of shotgun metagenomic data covering 11 diseases identified high microbial similarity between Crohn's disease and ulcerative colitis, Crohn's disease and colorectal cancer, Parkinson's disease and type 2 diabetes, and schizophrenia and type 2 diabetes [6]. It also found strong inverse correlations in Alzheimer's disease versus Crohn's disease and ulcerative colitis [6].

Metagenomics has also been used to identify microbial markers for disease diagnosis. A study of 1,012 subjects identified a novel faecal bacterial marker from a Lachnoclostridium species that was significantly enriched in colorectal adenoma [9]. The marker performed better than faecal immunochemical testing for detecting adenoma, with sensitivities for non-advanced and advanced adenomas of 44.2% and 50.8% at 79.6% specificity, compared to 0% and 16.1% for FIT at 98.5% specificity [9].

Environmental and Agricultural Monitoring

Metagenomics is used to monitor microbial communities in soil, water, and agricultural systems. It can assess the impact of management practices on microbial diversity and function. For example, metagenomics has been used to monitor the effect of raw versus digested manure on microbial diversity in anaerobic digestion of Napier grass [26]. It has also been applied to enhance acidogenic gas utilization in two-stage co-digestion via biogas recirculation [25].

In plant systems, metagenomics can profile leaf-associated microbial communities. A study of sagebrush leaf microbiomes used host genomic sequencing data to investigate the metagenomes of leaf-associated microbes, reconstructing two high-quality MAGs from greenhouse-grown plants and revealing that wild field-collected samples were dominated by Klebsiella and Aureobasidium species [14].

Combining Genomics and Metagenomics

Multi-Omics Integration

Genomics and metagenomics are complementary, and their integration can provide insights that neither approach achieves alone. Multi-omics analysis combining host genomics, metagenomics, and metabolomics has been used to explain site-specific differences in colon cancer. A study of 494 participants found unique profiles of the intestinal microbiome, metabolome, and host genome between right-sided and left-sided colon cancer [7]. The bacteria Flavonifractor plautii and Fusobacterium nucleatum, the metabolite L-phenylalanine, and the host genes PHLDA1 and WBP1 were key omics features of right-sided colon cancer, whereas Bacteroides sp. A1C1 and Parvimonas micra, the metabolites L-citrulline and D-ornithine, and the host genes TCF25 and HLA-DRB5 were dominant in left-sided colon cancer [7].

Host Depletion and Metagenome-Assembled Genomes

When metagenomic sequencing is performed on host-associated samples, the data contain both host and microbial reads. Host sequences can be removed by mapping to reference genomes. In the sagebrush leaf microbiome study, reads were mapped to the reference genomes of Artemisia tridentata, Artemisia annua, and the human reference genome to remove plant host and human-associated sequences before microbial analysis [14].

The recovered MAGs can then be analyzed for functional potential. In the Gambian pneumonia study, MAGs were used to assess antimicrobial resistance genes, revealing that the resistomes were highly species specific with more than 70% of detected AMR genes found exclusively in a single species [17].

Liquid Biopsy and Circulating Microbial DNA

Metagenomic sequencing of circulating microbial DNA in blood is an emerging approach for cancer detection. A study developed a liquid biopsy assay based on circulating microbial DNA for early detection of esophageal adenocarcinoma and high-grade dysplasia [11]. Using metagenomic sequencing, the study identified significant differences in microbial diversity and composition between disease groups, and a 6-marker panel achieved an AUC of 0.93 in the training cohort and 0.91 for esophageal adenocarcinoma and 0.88 for high-grade dysplasia in an independent testing cohort [11].

Practical Workflow Selection

Step 1: Define the Biological Question

The first decision is whether the question concerns a single organism or a community. If the question is about one species, such as identifying mutations in a bacterial isolate or characterizing a host genome, genomics is appropriate. If the question is about which organisms are present, their relative abundances, or their functional potential, metagenomics is appropriate.

Step 2: Assess Sample Characteristics

Consider the nature of the sample. Is it a pure culture or a mixed community? Is the target organism abundant or rare? Is there significant host contamination? For clinical specimens, is the patient on antibiotics? These factors influence the choice of sequencing approach and depth.

Step 3: Evaluate Reference Database Completeness

Read-based metagenomics relies on reference databases. If the community contains many novel or uncharacterized species, assembly-based metagenomics with MAG reconstruction is necessary. The completeness of the reference database should be assessed before choosing the analysis strategy.

Step 4: Consider Turnaround Time and Cost

For clinical applications, turnaround time is critical. tNGS can provide results in 10 to 16 hours at a fraction of the cost of mNGS [5]. For research applications, the cost of deep sequencing for strain-level resolution may be justified. The choice between short-read and long-read platforms also affects cost and turnaround time.

Step 5: Plan for Validation

Metagenomics results should be validated with orthogonal methods. For pathogen detection, positive results can be confirmed with targeted PCR or culture. For microbiome studies, key findings can be validated with quantitative PCR. The Lachnoclostridium marker for colorectal adenoma was identified by metagenomics and validated by targeted quantitative PCR [9].

Records and Measurements

Sequencing Quality Metrics

Standard quality metrics for sequencing data include read length, read depth, base quality scores, and GC content. For metagenomics, additional metrics include the proportion of reads that map to the host genome, the proportion of reads that map to reference databases, and the estimated community diversity.

Assembly Quality Metrics

For genomics, assembly quality is assessed by N50, L50, completeness, and contamination. For metagenomics, MAG quality is assessed by completion and contamination estimates based on single-copy marker genes. The MIMAG standards define high-quality MAGs as those with more than 90% completion and less than 5% contamination.

Reproducibility Records

Reproducibility requires documenting the software versions, parameters, and reference databases used in the analysis. Workflow management systems such as Nextflow and Snakemake support reproducibility. The TOFU-MAaPO pipeline provides a single-command workflow for large-scale metagenomic analysis [13]. Containerization with Docker ensures that the analysis environment is consistent across runs.

Common Failure Patterns

Host Contamination

Host DNA can dominate metagenomic sequencing data, reducing the effective depth for microbial reads. This is a common problem in clinical specimens and plant-associated samples. Host depletion methods, such as differential lysis or probe-based capture, can reduce host contamination before sequencing.

Uneven Community Abundance

Metagenomic samples often contain a few dominant species and many rare species. Deep sequencing is required to detect rare organisms, but this increases cost. The choice of sequencing depth should balance the need to detect rare species against the cost of sequencing.

Reference Database Bias

Read-based taxonomic assignment is biased toward organisms represented in reference databases. Novel species may be misclassified or missed entirely. Assembly-based approaches with MAG reconstruction can overcome this limitation but require more computational resources.

DNA Extraction Bias

DNA extraction methods differ in their ability to lyse different cell types. Gram-positive bacteria, fungi, and viruses require different lysis conditions. The choice of extraction method can bias the observed community composition. The comparison of six DNA extraction methods for Nanopore sequencing found significant differences in yield and detection accuracy [18].

Antibiotic Exposure

Prior antibiotic exposure reduces the sensitivity of culture but has less effect on mNGS. In a study of 511 clinical specimens, mNGS had a sensitivity of 52.5% versus 34.2% for culture in cases with antibiotic exposure [12]. However, antibiotic exposure can still affect the microbial community composition and the interpretation of results.

Limitations and Interpretation

Detection Thresholds

Metagenomics has detection thresholds that depend on sequencing depth and community complexity. Low-abundance organisms may fall below the detection limit. The limit of detection for tNGS assays was 50 to 450 CFU/mL in the lower respiratory tract infection study [5]. These thresholds should be considered when interpreting negative results.

Contamination and Background Noise

Metagenomic data can contain contamination from reagents, laboratory environments, and the host. Negative controls are essential to identify and subtract background contamination. The interpretation of low-abundance organisms should be cautious, particularly in clinical specimens.

Functional Inference

The presence of a gene in a metagenome does not confirm that the gene is expressed or that the organism is metabolically active. Functional annotation provides potential, not proof. Metatranscriptomics and metabolomics can provide complementary evidence of activity.

Strain-Level Resolution

Read-based metagenomics cannot resolve strain-level variation. Assembly-based approaches with MAGs can distinguish strains, but this requires high sequencing depth and advanced methods. The genome unit number metric was developed to quantify within-species diversity based on MAGs [16].

Safety and Regulatory Context

Data Sharing and Privacy

Genomic and metagenomic data may contain identifiable human information. The NIH Genomic Data Sharing Policy establishes expectations for data sharing, including informed consent, privacy protection, and data use limitations [3]. Researchers should review and comply with applicable policies before depositing or sharing data.

Data Management Standards

The FAIR Guiding Principles describe the characteristics that data resources should have to facilitate reuse, including findability, accessibility, interoperability, and reusability [4]. Applying these principles to genomic and metagenomic data supports reproducibility and secondary analysis.

Data Repositories

Public data repositories such as NCBI provide access to genomic and metagenomic data [2]. The Sequence Read Archive contains over 600,000 metagenomes that can be used for secondary analysis [13]. EMBL-EBI Training provides educational resources for bioinformatics and data management [1].

Standards Development

Community standards for data annotation and exchange are under active development. The RCN4GSC meeting report describes efforts to establish a testbed for managing data at the interface of biodiversity and genomics and metagenomics, including an element-by-element comparison of the Darwin Core and GSC MIxS standards [21]. These standards support interoperability across data types and communities.

Professional Escalation Criteria

When to Consult a Specialist

Researchers should consider consulting a bioinformatics specialist or clinical microbiologist when the analysis requires specialized expertise. This includes cases where the community composition is highly complex, where novel species are suspected, or where the clinical interpretation of results is uncertain.

When to Use a Clinical Reference Laboratory

For clinical applications, results from metagenomic sequencing should be confirmed in a Clinical Laboratory Improvement Amendments certified laboratory when the result will guide patient management. The diagnostic performance of mNGS varies by specimen type and clinical context, and results should be interpreted in the context of the full clinical picture.

When to Escalate for Regulatory Review

Research involving human subjects, including genomic and metagenomic studies of human samples, may require institutional review board approval. Studies involving pathogens may require biosafety review. Researchers should consult their institutional compliance offices before initiating studies with potential regulatory implications.

Frequently Asked Questions

What is the main difference between genomics and metagenomics?

Genomics studies the complete genetic material of a single organism, such as a bacterial isolate or a human patient. Metagenomics studies the collective genetic material of all organisms in a mixed community sample, such as feces, soil, or a clinical specimen. The choice depends on whether the biological question concerns one organism or a community.

Can metagenomics replace genomics?

Metagenomics cannot replace genomics for questions that require a complete, high-quality genome of a single organism. Genomics provides the depth and completeness needed for variant calling, structural variant detection, and comparative genomics. Metagenomics is superior for questions about community composition, unculturable organisms, and functional potential of mixed communities.

When should I use targeted next-generation sequencing instead of shotgun metagenomics?

Targeted next-generation sequencing is appropriate when the suspected pathogens are known and can be included in a panel. It is faster and less expensive than shotgun metagenomics, with test costs reduced to a quarter or half of mNGS [5]. However, tNGS only detects organisms in the targeted panel and may miss unexpected pathogens.

How does antibiotic exposure affect metagenomic pathogen detection?

Prior antibiotic exposure reduces the sensitivity of culture but has less effect on metagenomic next-generation sequencing. In a study of 511 clinical specimens, mNGS had a sensitivity of 52.5% versus 34.2% for culture in cases with antibiotic exposure [12]. This makes mNGS valuable for patients who have already received empirical antibiotics.

What are metagenome-assembled genomes and why are they important?

Metagenome-assembled genomes are genomes reconstructed from metagenomic sequencing data by grouping contigs that likely originate from the same organism. They enable the discovery of novel species and the analysis of unculturable organisms. A study of the Estonian Microbiome Cohort reconstructed 84,762 MAGs representing 2,257 species, including 353 previously uncharacterized species [16].

How do I choose between short-read and long-read sequencing for metagenomics?

Short-read sequencing is cost-effective and accurate but produces fragmented assemblies. Long-read sequencing resolves repetitive regions and produces more complete genomes. The myloasm assembler for modern long reads assembled three times more complete circular contigs than the next-best assembler on real-world ONT metagenomes [15]. The choice depends on the research question and budget.

What quality metrics should I report for metagenome-assembled genomes?

Report completion and contamination estimates based on single-copy marker genes, along with the number of contigs, N50, and total length. High-quality MAGs typically have more than 90% completion and less than 5% contamination. The TOFU-MAaPO workflow integrates multiple binning tools with a unified refinement strategy to improve MAG quality [13].

How can I validate metagenomic findings?

Validate key findings with orthogonal methods such as targeted quantitative PCR, culture, or independent sequencing. The Lachnoclostridium marker for colorectal adenoma was identified by metagenomics and validated by targeted quantitative PCR [9]. For clinical applications, confirm results in a certified laboratory when the result will guide patient management.

Related Bioinformatics Guides

• Ethical Considerations in Computational Genomics
• Deep Learning for Functional Genomics
• The Rise of Omics: Genomics, Proteomics, and Metabolomics
• Pangenome Graph Construction for Bacterial Genomics
• The UK Biobank: Managing Massive Biological Datasets

References and Further Reading

• EMBL-EBI Training. European Bioinformatics Institute.
• NCBI Data Resources. National Center for Biotechnology Information.
• Genomic Data Sharing Policy. National Institutes of Health.
• The FAIR Guiding Principles. Scientific Data.
• Enhancing lower respiratory tract infection diagnosis: implementation and clinical assessment of multiplex PCR-based and hybrid capture-based targeted next-generation sequencing.. EBioMedicine, 2024.
• Meta-analysis of the human gut microbiome uncovers shared and distinct microbial signatures between diseases.. mSystems, 2024.
• Distinct microbes, metabolites, and the host genome define the multi-omics profiles in right-sided and left-sided colon cancer.. Microbiome, 2024.
• Effect of Metagenomic Next-Generation Sequencing on Clinical Outcomes of Patients With Severe Community-Acquired Pneumonia in the ICU: A Multicenter, Randomized Controlled Trial.. Chest, 2025.
• A novel faecal Lachnoclostridium marker for the non-invasive diagnosis of colorectal adenoma and cancer.. Gut, 2020.
• Restoration of the human skin microbiome following immune recovery after hematopoietic stem cell transplantation.. Cell host & microbe, 2025.
• A machine-learning informed circulating microbial DNA signature for early diagnosis of esophageal adenocarcinoma.. Gut microbes, 2026.
• Microbiological Diagnostic Performance of Metagenomic Next-generation Sequencing When Applied to Clinical Practice.. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America, 2018.
• TOFU-MAaPO: fast, scalable and reproducible analysis of large metagenome sequence data from the Sequence Read Archive.. 2026.
• Exploring sagebrush leaf microbial metagenomes from deep, host-derived sequencing.. 2026.
• High-resolution metagenome assembly for modern long reads with myloasm.. 2026.
• Metagenome-assembled genomes from a population-based cohort uncover novel gut species and within-species diversity, revealing prevalent disease associations.. 2026.
• Shotgun metagenomic profiling of bacterial microbiomes, metagenome-assembled genomes and antimicrobial resistance in respiratory and blood samples from Gambian children with pneumonia. 2026.
• Comparison of 6 DNA extraction methods for isolation of high yield of high molecular weight DNA suitable for shotgun metagenomics Nanopore sequencing to detect bacteria. BMC Genomics, 2023.
• The fifth international hackathon for developing computational cloud-based tools and resources for pan-structural variation and genomics. F1000Research, 2024.
• Metagenomics Comparison of Buruli and Non Buruli Ulcer Skin Wound. Journal of Clinical and Diagnostic Research, 2022.
• RCN4GSC Meeting Report: Initiating a Testbed for Managing Data at the Interface of Biodiversity and Genomics/Metagenomics, May 2011. Standards in Genomic Sciences, 2012.
• From Genomics to Metagenomics in the Era of Recent Sequencing Technologies.. Methods in molecular biology, 2023.
• A comparison of classification methods for gene prediction in metagenomics. 2014.
• Metagenomics approach for Polymyxa betae genome assembly enables comparative analysis towards deciphering the intracellular parasitic lifestyle of the plasmodiophorids.. Genomics, 2021.
• Enhanced acidogenic gas utilization in two-stage co-digestion via biogas recirculation: Metagenomics analysis. Renewable Energy, 2026.
• Combination of flow cytometry and metagenomics to monitor the effect of raw vs digested manure on microbial diversity in anaerobic digestion of Napier grass. Environmental Monitoring and Assessment, 2025.
• Population genomics and antimicrobial resistance dynamics of Escherichia coli in wastewater and river environments. Communications Biology, 2021.
• Comparative Respiratory Tract Microbiome Between Carbapenem-Resistant Acinetobacter baumannii Colonization and Ventilator Associated Pneumonia. Frontiers in Microbiology, 2022.
• Stakeholder consultation insights on the future of genomics at the clinical-public health interface. Translational Research, 2014.

This article is educational and does not replace validated analysis plans, institutional policy, clinical interpretation, or specialist review.