Long-Read Metagenomics Sequencing — Complete Microbial Genome Recovery & Functional Analysis by PacBio & Nanopore

Comprehensive metagenomic sequencing on PacBio HiFi (Revio) and Oxford Nanopore PromethION — delivering complete metagenome-assembled genomes (cMAGs), strain-level taxonomic resolution, and functional annotation for human microbiome research, environmental microbiology, and industrial applications

CD Genomics provides long-read metagenomics sequencing on PacBio Revio HiFi and ONT PromethION platforms — delivering complete metagenome-assembled genomes (cMAGs), strain-level taxonomic resolution, and comprehensive functional annotation including antibiotic resistance profiling, biosynthetic gene cluster discovery, and metabolic pathway reconstruction from complex microbial communities.

The vast majority of microbial diversity remains inaccessible to culture-based methods — estimates suggest that over 70% of bacterial and archaeal species have never been isolated in the laboratory. Metagenomic sequencing bypasses this limitation by extracting and sequencing DNA directly from environmental or host-associated microbial communities, providing a window into the uncultured majority. However, the short-read sequencing platforms that have dominated metagenomics for the past two decades produce reads of 150–300 bp, which are insufficient to assemble complete microbial genomes from complex communities — most short-read metagenome-assembled genomes (MAGs) remain fragmented, contaminated, and lacking the genomic resolution needed for strain-level analysis, gene-synteny studies, and accurate functional annotation.

Long-read metagenomics has transformed this landscape. Platforms from Pacific Biosciences (HiFi on Revio) and Oxford Nanopore Technologies (PromethION) produce reads of 15–100+ kilobases directly from microbial community DNA, enabling the recovery of complete, circular metagenome-assembled genomes (cMAGs) at unprecedented resolution. Recent benchmark studies have demonstrated that long-read metagenomics yields 44–64× more complete MAGs per gigabase of sequencing data compared to short-read approaches, and that PacBio HiFi, in particular, achieves the highest cost-efficiency for finished microbial genomes at approximately $16 per cMAG — compared to $95–$2,370 per cMAG for Illumina. For complex microbial communities, long-read metagenomics is no longer a niche complement to short-read sequencing but the definitive approach for achieving genome-resolved metagenomics.

At CD Genomics, we provide long-read metagenomics sequencing services on both PacBio HiFi (Revio) and Oxford Nanopore PromethION platforms, with flexible coverage, multiplexed barcoding, and comprehensive bioinformatics pipelines covering cMAG assembly, taxonomic profiling, functional annotation, and downstream biological interpretation. Whether your research targets the human gut microbiome, environmental microbial ecology, or industrial bioprospecting, our team will design and execute a metagenomics strategy tailored to your research objectives.

Why Choose Our Long-Read Metagenomics Service

• Complete Microbial Genomes from Complex Communities — Long reads (15–100+ kb) span repetitive elements and genomic rearrangements to produce complete, circular MAGs at strain-level resolution — including near-identical strains (≥99.9% ANI) that cannot be resolved by short-read approaches
• Dual-Platform Flexibility — PacBio HiFi (Q30+ consensus accuracy, metaMDBG assembly) for the highest quality cMAGs; ONT ultra-long reads (metaFlye assembly) for spanning the longest genomic elements; or hybrid strategies combining both platforms for the most comprehensive community genome recovery
• Proven Cost-Effectiveness for Finished Genomes — Published benchmarks demonstrate that long-read metagenomics delivers 44–64× more complete MAGs per Gbp than Illumina, with PacBio HiFi at ~$16/cMAG representing the most cost-efficient route to finished microbial genomes at scale
• Strain-Level Taxonomic & Functional Resolution — Complete circular genomes enable species-level and strain-level taxonomic classification, accurate gene catalog construction, mobile genetic element tracking, antibiotic resistance gene (ARG) characterization, and horizontal gene transfer detection that is impossible from fragmented short-read assemblies
• Integrated Functional & Metabolic Annotation — From CAZy carbohydrate-active enzymes and antiSMASH biosynthetic gene clusters to KEGG pathway mapping, eggNOG functional annotation, and custom metabolic model reconstruction — our bioinformatics pipelines extract the full functional potential from every assembled genome
• End-to-End Project Management — From sample collection guidance and DNA extraction optimisation through sequencing, cMAG assembly, annotation, and publication-ready reporting — a dedicated project scientist manages every stage with regular progress updates

What Is Long-Read Metagenomics Sequencing — and Why Long Reads Are Essential for Complete Microbial Genomes

Metagenomic sequencing is the process of extracting and sequencing DNA directly from an entire microbial community — from soil or ocean water to the human gut — without the need to culture individual microbial species. The goal of metagenomics is to reconstruct the genomes of every microorganism present in the sample and to understand their functional potential, evolutionary relationships, and interactions within the community. For the vast majority of microbial species that remain unculturable, metagenomics is the only way to access their genomes and biology.

Short-read metagenomic sequencing (Illumina, 150–300 bp paired-end reads) has been the dominant approach for over a decade and has produced thousands of reference MAGs. However, the fundamental limitation of short reads is that they cannot span repetitive elements, resolve closely related strains, or produce complete circular genomes from complex communities. Most short-read MAGs are assembled as collections of hundreds or thousands of contigs, with contamination from co-assembled strains and fragmented gene models that compromise downstream functional interpretation. The Human Microbiome Project 2 and similar large-scale efforts have demonstrated that short-read metagenomics recovers vanishingly few complete, circular MAGs from human stool samples — and essentially zero from complex environmental communities.

Long-read metagenomics solves this problem at its root. A single PacBio HiFi read of 15–25 kb or an ONT read of 50–100+ kb can span entire rRNA operons, mobile genetic elements, phage integration sites, and strain-specific genomic islands — producing complete, circular chromosomes from individual microbial species. When applied to human stool metagenomes, long-read sequencing routinely recovers dozens of complete MAGs per sample, including near-identical strains (≥99.9% average nucleotide identity) that collapse into a single bin in short-read assemblies. For environmental metagenomics, long reads provide access to the complete genomes of low-abundance species and resolve the genomic heterogeneity that defines microbial population structure.

At CD Genomics, our long-read metagenomics service is designed around this fundamental advantage — delivering finished microbial genomes that enable the full spectrum of downstream analyses, from taxonomic profiling and functional annotation to pangenomics, population genetics, and biosynthetic gene cluster discovery.

Long-Read Metagenomics Applications — From Human Microbiome to Environmental Surveillance

Beyond human microbiome and environmental applications, long-read metagenomics is increasingly applied in time-series and longitudinal study designs that track microbial community dynamics — including strain-level monitoring during disease progression, therapeutic interventions, and environmental perturbations — where the ability to recover complete genomes at each timepoint enables direct observation of within-strain genomic evolution, horizontal gene transfer events, and population genetic changes that fragmented short-read MAGs cannot resolve. Metatranscriptomic integration and spatial metagenomics represent additional emerging frontiers where long-read platforms provide simultaneous access to both community genomic potential and active functional expression from the same samples.

Human Gut Microbiome Research

• Complete genome recovery from longitudinal stool cohorts — long-read metagenomics recovers dozens of cMAGs per sample, enabling genome-resolved analysis of microbiome dynamics over time, in response to diet, disease, or therapeutic interventions
• Strain tracking and within-host evolution — complete circular genomes provide the nucleotide-level resolution needed to track strain transmission, within-host genomic evolution, and the emergence of antibiotic resistance in gut commensals and pathobionts
• Microbiome-phenotype association studies — cMAGs enable gene-level genome-wide association studies (mGWAS) linking specific microbial genes and pathways to host phenotypes, including metabolic disease, immune-mediated disorders, and drug response variability

Environmental Microbiology

• Soil and rhizosphere metagenomics — complete genomes from soil microbial communities reveal the functional potential for nutrient cycling, plant growth promotion, bioremediation, and carbon sequestration that is inaccessible from fragmented short-read assemblies
• Marine and freshwater microbiology — long-read metagenomics recovers complete genomes from aquatic microbial communities, including low-abundance species and uncultured lineages that constitute the microbial rare biosphere
• Extreme environment microbiology — finished genomes from hot springs, deep-sea vents, and polar environments enable comparative genomics of extremophiles and the discovery of novel thermostable enzymes and metabolic pathways

Industrial & Biotechnological Applications

• Bioprospecting and natural product discovery — complete genomes enable comprehensive biosynthetic gene cluster (BGC) discovery via antiSMASH, BiG-SCAPE, and related tools, revealing the full secondary metabolite potential of environmental microbiomes for drug discovery and biotechnology
• Wastewater treatment and bioreactor metagenomics — tracking microbial community composition and functional dynamics in engineered systems, including antibiotic resistance gene surveillance and the characterization of key functional guilds for process optimization
• Industrial enzyme discovery — complete genes and operons from cMAGs provide intact, full-length enzyme sequences for CAZy, MEROPS, and other industrially relevant enzyme families, directly supporting synthetic biology and enzyme engineering pipelines

Agricultural & Food Microbiology

• Rumen and livestock microbiome — complete MAGs from the gastrointestinal microbiomes of production animals enable understanding of feed efficiency, methane emissions, and the impact of dietary interventions on microbial community structure and function
• Food safety and spoilage microbiology — strain-resolved genomic surveillance of foodborne pathogens and spoilage organisms in production environments, including tracking of antimicrobial resistance determinants
• Plant microbiome engineering — complete genomes of plant-growth-promoting rhizobacteria and their antagonists support the design of synthetic microbial communities for sustainable agriculture

Our Bioinformatics Delivers cMAG Assembly, Strain-Level Taxonomic Profiling, and Multi-Database Functional Annotation from Long-Read Metagenomic Data

Our metagenomics bioinformatics pipeline supports the complete analysis workflow from raw read processing through cMAG assembly, taxonomic and functional annotation, and advanced biological interpretation — organized across two service tiers to accommodate different research objectives and budgets.

Sample Input Requirements for Long-Read Metagenomics — DNA Quality, Coverage Targets, and Multiplexing Options by Community Type

Case Study: Culture-Independent Meta-Pangenomics by Long-Read Metagenomics Reveals Associations with Pediatric Undernutrition

Minich JJ, Allsing N, Din MO, Tisza MJ, Maleta K, McDonald D, Hartwick N, Mamerto A, Brennan C, Hansen L, Shaffer J, Murray ER, Duong T, Knight R, Stephenson K, Manary MJ, Michael TP. Culture-independent meta-pangenomics enabled by long-read metagenomics reveals associations with pediatric undernutrition. Cell. 2025;188(23):6666-6686.e25.

1. Background

Pediatric undernutrition (stunting and wasting) affects millions of children worldwide and is associated with alterations in the developing gut microbiome. However, previous short-read metagenomic studies of the gut microbiome in undernutrition have been limited by their inability to recover complete microbial genomes from the study cohort — meaning that strain-level dynamics, gene content variation, and pangenomic diversity have remained largely unexplored. The authors hypothesized that long-read metagenomics could recover complete, circular MAGs (cMAGs) from a longitudinal pediatric cohort and enable genome-resolved analysis of microbial population dynamics associated with linear growth faltering.

2. Methods

The study sequenced 47 fecal samples from 8 Malawian children (aged 12–48 months) across multiple time points using three platforms: PacBio Revio (HiFi CCS, 591 Gbp total), Oxford Nanopore PromethION (R10.4.1, 1,172 Gbp total), and Illumina NovaSeq X (625 Gbp total). Assembly was performed using platform-optimised approaches: metaMDBG for PacBio HiFi, metaFlye for ONT, and MEGAHIT/metaSPAdes for Illumina. cMAGs were quality-filtered (CheckM2, ≥90% completeness, ≤5% contamination), taxonomically classified (GTDB-Tk), and used for pangenome construction and metagenomic GWAS against clinical phenotypes. The resulting cMAG database was applied to an expanded cohort of 210 samples from 42 participants.

3. Results

Figure 2. Comparison of PacBio HiFi, ONT, and Illumina sequencing for human gut metagenomics. Long-read platforms (PacBio HiFi + ONT) produced 986 cMAGs (839 circular) from 47 samples — 44–64× more cMAGs per Gbp than Illumina, which recovered zero complete MAGs. PacBio HiFi achieved the highest cMAG quality and cost-efficiency (~$16/cMAG). Adapted from Minich et al. (2025), Cell, CC BY 4.0.

Key Findings

• Platform comparison — cMAG yield and quality: Long-read platforms dramatically outperformed Illumina for complete genome recovery. PacBio HiFi assembled ~23.4 cMAGs per sample at 30 Gbp (vs ~10.1 for ONT), with significantly higher completeness and lower contamination. Illumina recovered zero complete circular MAGs from any sample — producing only fragmented, contaminated bins unsuitable for pangenome analysis. The cost per cMAG was ~$16 for PacBio HiFi, ~$25 for ONT, and $95–$2,370 for Illumina.
• Complete genome database: Combined PacBio HiFi + ONT hybrid assembly (metaMDBG) produced 986 cMAGs representing 363 species (74 putative novel species at ANI <95%), including 839 circular chromosomes. This cMAG database reduced unclassified metagenomic reads from 52.3% to 16.5% when integrated into the GTDB reference — a 68.5% improvement in taxonomic assignment.
• Clinical association — genome instability and growth faltering: The study identified a significant correlation between gut microbial genome instability — measured as decreasing average nucleotide identity (ANI) within species over time — and declining length-for-age Z score (LAZ), the clinical measure of linear growth faltering in children. Machine learning models identified specific microbial species whose genomic dynamics predicted growth trajectories.
• Metagenomic GWAS (mGWAS) and functional associations: Pangenome-wide association analysis identified 89 bacterial genes significantly associated with improvements in LAZ, including arnC (involved in LPS modification, potentially modulating host immune interactions). Species from Bifidobacterium, Megasphaera, Faecalibacterium, and Prevotella showed gene-level associations with growth outcomes and breastfeeding status. Higher phage integration frequency in Faecalibacterium in children with improving LAZ suggested that viral-bacterial interactions may influence growth trajectories.

4. Conclusions

This landmark study demonstrated that long-read metagenomics — particularly PacBio HiFi — enables the recovery of hundreds of complete, circular microbial genomes from a modest number of clinical samples, providing the genome-resolved resolution needed to discover biologically and clinically meaningful associations that are invisible to short-read approaches. The finding that gut microbial genome instability correlates with pediatric growth faltering opens new avenues for understanding the role of the microbiome in undernutrition and highlights the unique value of long-read metagenomics for generating the complete genomic data required for pangenome-wide association studies in human populations.

Interpretation Boundaries for Long-Read Metagenomic Analysis

• cMAG recovery rates are community-dependent and not guaranteed for all species. While long-read metagenomics dramatically outperforms short-read for complete genome recovery, the number and quality of cMAGs depend on community evenness, species richness, and the abundance of individual community members. Very low-abundance species (<0.1% relative abundance) and species with extremely high genomic complexity or strain heterogeneity may still yield incomplete MAGs. Expected cMAG counts are estimated during project design based on published benchmarks for similar community types
• Strain-level resolution has fundamental limits at very high strain diversity. While long reads resolve near-identical strains better than short reads, communities containing dozens of nearly identical strains (≥99.99% ANI) — common in some environmental and host-associated microbiomes — may still show strain-level assembly ambiguity. Per-sample strain diversity is assessed during analysis and documented in project reports
• Functional annotation quality depends on database completeness and homology. Gene function prediction from assembled MAGs relies on alignment to reference databases (KEGG, CAZy, CARD, etc.). Novel genes, deeply divergent lineages, and taxa with few cultured representatives may have low functional annotation rates despite complete genome sequences. Annotation rates are reported for each MAG
• Metagenomic data are for research use only. All sequencing data, assembled genomes, and functional annotations are generated for research applications and are not validated for clinical diagnosis, treatment decisions, or clinical variant interpretation in individual patients or specimens
• Cross-platform and cross-study comparability should be evaluated. Different sequencing platforms, assembly algorithms, and bioinformatics pipelines produce systematically different MAG sets and taxonomic profiles. Direct comparison of results generated by different methods requires appropriate normalization and cross-validation. For longitudinal or multi-cohort studies, we recommend consistent use of the same platform and bioinformatics pipeline throughout

Metagenomics Deliverables Include Raw Data, Complete cMAGs, Functional Annotations, and Publication-Ready Reports

Deliverable Examples for Long-Read Metagenomics Sequencing Projects

1. Raw sequencing data files (FASTQ, POD5 or BAM) with basecalling quality metrics, read length distributions, and sequencing depth statistics per sample.

2. Assembly and MAG data: assembled contigs (FASTA), MAG bins with quality metrics (completeness, contamination, strain heterogeneity from CheckM2), circular cMAG sequences with predicted replication origin and termini, and MAG classification (GTDB-Tk taxonomy).

3. Metagenomics analysis report including: community composition profiles (phylum through species level), alpha and beta diversity analyses, cMAG quality summary table with taxonomic assignments, functional annotation tables (KEGG, CAZy, antiSMASH, eggNOG), antibiotic resistance gene profiles, and comparative analysis results.

4. Publication-ready figures, interactive Krona taxonomic visualizations, Circos genome maps for key cMAGs, and NCBI BioProject/SRA submission support for data release.

Figure 3. Representative deliverable formats for long-read metagenomics projects. Left: cMAG quality summary and taxonomic classification. Center: community taxonomic composition and diversity analysis. Right: functional annotation and antibiotic resistance gene profiling. AI-generated representative data.