Microbial Whole-Genome De Novo Sequencing — Complete Long-Read Genome Assembly & Annotation

Short-read assemblies leave gaps, collapse repeats, and routinely miss plasmids — producing fragmented drafts that undermine downstream comparative genomics, AMR surveillance, and publication-quality analysis. CD Genomics' microbial whole-genome de novo sequencing service solves this by combining PacBio HiFi and Oxford Nanopore (ONT) long-read sequencing to deliver closed, chromosome-level assemblies with all plasmids resolved — not draft genomes, but finished references ready for submission and analysis.

Our dual-platform approach serves academic microbiology labs, public health agencies, and biotech teams requiring complete bacterial, fungal, and viral genome assemblies. With over a decade of experience in microbial genomics with long-read sequencing, we provide end-to-end support — from HMW DNA extraction guidance through assembly, polishing, annotation, and comparative bioinformatics.

What you get with our microbial de novo sequencing service

• Closed, chromosome-level genome assemblies — zero gaps, all plasmids resolved
• Dual-platform PacBio HiFi and ONT PromethION — platform choice optimised for your organism
• Complete bioinformatics pipeline — assembly, polishing, annotation, comparative genomics
• HMW DNA expertise — extraction QC guidance for GC-rich, repeat-dense, and challenging microbes
• Publication-ready reporting with assembly QC metrics (N50, genome completeness, BUSCO scores)

Why Long-Read Sequencing Transforms Microbial De Novo Assembly

Short-read platforms like Illumina have driven microbial genomics for over a decade, but their fundamental limitation — reads of 150–300 bp — creates persistent assembly challenges. Bacterial genomes typically contain ribosomal RNA operons (5–7 kb), insertion sequence (IS) elements (0.7–2 kb), prophage regions (10–100 kb), and near-identical repeat families that exceed single read length. When an assembler encounters a repeat longer than the read, it cannot determine which copy belongs to which genomic location, producing fragmented assemblies with dozens to hundreds of contigs.

The assembly metrics tell the story. A typical Illumina-only assembly of a 5 Mb bacterial genome yields 50–200 contigs, an N50 of 80–150 kb, and routinely misses one or more plasmids. In contrast, PacBio SMRT sequencing with HiFi reads (10–25 kb, >99.9% accuracy) or ONT ultralong reads (up to 100 kb+) can resolve the same genome to 1–3 contigs — often a single circular chromosome plus individual plasmid contigs. This is not incremental improvement; it is a qualitative shift from draft to finished genome status [1].

With a closed genome, you can confidently map structural variants, characterise complete plasmid sequences, resolve IS element positions, identify prophage integration sites, and perform gene neighbourhood analysis without ambiguity about whether adjacent genes are truly adjacent or separated by an assembly break. For labs publishing microbial genomes, submitting complete assemblies to GenBank satisfies increasingly stringent reviewer requirements that prefer closed reference sequences.

Complete Service Workflow — From HMW DNA to Closed Genome

Every microbial de novo project follows a validated six-stage pipeline, from sample receipt to delivery of the finished, annotated genome. Each stage includes defined QC checkpoints; no sample proceeds to the next stage without meeting quality thresholds.

Microbial Genomes We Support — Bacteria, Fungi, and Viruses

Our microbial de novo sequencing workflows have been applied across the taxonomic spectrum, from small bacterial genomes (~1 Mb) to large fungal genomes (>100 Mb). Each organism type presents distinct assembly challenges; our protocols and platform recommendations are tailored accordingly.

Bacteria. Bacterial genomes (typically 2–10 Mb) are the most common targets for de novo sequencing, and PacBio HiFi routinely closes them in a single run. We have extensive experience with both model organisms (E. coli, Bacillus, Pseudomonas) and non-model environmental isolates, including GC-rich Actinobacteria, AT-rich Mollicutes, and multi-plasmid Enterobacteriaceae. Our bacterial whole-genome de novo sequencing page provides organism-specific details.

Fungi. Fungal genomes (10–100+ Mb) present challenges beyond bacteria: larger size, higher repeat content (transposable elements, telomeric repeats, centromeres), and more complex gene structures with introns. Long-read sequencing is essential for resolving these features. ONT ultralong reads are often recommended for spanning large repetitive regions, while PacBio HiFi provides the accuracy needed for gene annotation. See our fungal whole-genome de novo sequencing service for detailed protocols.

Viruses. Viral genomes (typically 5–200 kb) benefit from long-read sequencing for resolving genomic termini, repeat regions, and mixed-population quasispecies analysis. Direct sequencing of viral DNA or cDNA without amplification eliminates PCR bias and enables detection of minority variants. Our viral genome de novo sequencing service covers both DNA and RNA viruses with appropriate library preparation strategies.

Why CD Genomics for Microbial De Novo Sequencing

Dual-Platform Without Platform Bias. We operate both PacBio Revio and ONT PromethION systems in-house, which means we recommend the right platform for your genome — not the one we happen to own. For projects where neither platform alone is sufficient, we design hybrid strategies that combine PacBio accuracy with ONT read length.

Closed-Genome Commitment. We do not consider a draft assembly acceptable. Every project is delivered as a closed, chromosome-level assembly — complete chromosomes circularized and verified, all plasmids resolved, zero gaps. If our initial assembly does not meet these standards, we invest additional sequencing and analysis until it does, at no extra cost to you.

Bioinformatics Depth, Not Just Base Calls. Assembly is where most CROs stop. We continue through polishing, circularization, annotation, and — when requested — comparative genomics, phylogenetics, and functional profiling. The deliverable is not a collection of FASTA files; it is a fully characterized genome ready for publication, GenBank submission, or downstream analysis.

QC Rigor at Every Stage. From HMW DNA integrity verification at intake through BUSCO completeness assessment at delivery, every stage has defined QC thresholds. Samples that fail QC are not silently processed — we contact you, explain the issue, and propose solutions before proceeding.

End-to-End Scientific Partnership. Our team includes molecular biologists, bioinformaticians, and microbiologists who understand the biological context of your project. We do not just run pipelines — we discuss experimental design, help interpret results, and support manuscript preparation. Whether you are sequencing a single clinical isolate or a 500-genome population study, we provide the same level of scientific engagement.

Demo Results & Data Visualization

Below are representative examples of the visualization outputs included in every microbial de novo sequencing project. These figures are generated from real assembly and annotation data and are designed for direct inclusion in publications, presentations, or grant applications.

Circular Genome Map

CGView circular genome map of a bacterial isolate assembled by PacBio HiFi. The outermost ring shows CDS on the forward strand; rings 2 and 3 show GC content and GC skew; inner rings display rRNA and tRNA positions. Plasmid contigs are displayed as separate circular maps in the final report.

Assembly Quality Comparison

Assembly quality comparison between Illumina-only (short-read), PacBio HiFi, ONT, and hybrid approaches for the same ~5 Mb bacterial genome. Bar charts show contig count (lower is better), N50 (higher is better), and BUSCO completeness (higher is better). The hybrid approach achieves the optimal balance of contiguity and accuracy.

Phylogenetic Tree from Comparative Genomics

Maximum-likelihood phylogenetic tree constructed from core-genome SNP alignment of 24 bacterial isolates. The tree is rooted at midpoint and bootstrap support values are shown at nodes. This analysis is included in the Advanced Bioinformatics Package and is suitable for publication as a supplementary figure.

For research use only. Not for use in diagnostic procedures.