
Every fragmented genome assembly tells the same story: the interesting parts — the repeats, the structural variants, the centromeres — are missing. When a PI submits a novel isolate for whole-genome sequencing and receives back 5,000 contigs with an N50 of 200 kb, the question is not whether data was generated, but whether the genome is actually complete. It is not. Short reads count genes; they cannot build genomes.
PacBio HiFi sequencing changes this equation. By combining single-molecule real-time (SMRT) detection with circular consensus sequencing (CCS), HiFi reads deliver Q30+ accuracy (>99.9%) at 15–25 kilobase lengths — accuracy that matches short reads and length that spans the repetitive elements that fragment Illumina assemblies. At CD Genomics, we operate the full PacBio platform suite — Revio, Vega, and Sequel IIe — and we have built 40+ application-specific services across human, animal, plant, microbial, transcriptomic, and epigenomic research. If your project demands complete, publication-grade genomes, full-length isoform discovery, or native methylation detection without bisulfite, PacBio HiFi is the platform — and we are here to help you deploy it.
Short-read sequencing was not designed to assemble genomes. It was designed to count reads. Every time an Illumina fragment encounters a segmental duplication longer than 300 bp, a GC-rich promoter, or a centromeric satellite array, the assembler has no way to resolve which copy goes where. The result is a fragmented draft — thousands of contigs, collapsed repeats, missing structural variants, and haplotype information lost to computational phasing guesswork.
PacBio HiFi reads solve this problem at the molecular level. With an average length of 15–25 kb and Q30+ per-base accuracy, a single HiFi read spans the typical repeat unit 50 to 100 times over. The assembler no longer needs to guess — it can walk across the repeat and anchor both sides unambiguously. This is not an incremental improvement. It is the difference between a draft and a finished genome.
The consequences are concrete. T2T genome assemblies — complete from telomere to telomere, including centromeres and ribosomal DNA arrays — are now routine with HiFi. Haplotype-resolved T2T assemblies phase both parental genomes independently, revealing allele-specific structural variation, gene conversion events, and imprinting patterns that short reads cannot access. For researchers in animal and plant de novo genome sequencing, HiFi routinely delivers chromosome-scale contiguity — N50 values exceeding 50 Mb — from a single Revio SMRT Cell.
HiFi accuracy also transforms variant detection. Unlike noisy long reads that require deep coverage and computational polishing to distinguish true variants from sequencing errors, HiFi reads call SNVs, indels, and structural variants directly — including variants in low-complexity regions that short-read aligners systematically mis-map. For human structural variation detection, HiFi captures the full variant spectrum in a single experiment: SNVs, indels, SVs >50 bp, and mobile element insertions — all haplotype-phased.
PacBio SMRT sequencing operates on a principle that is elegantly simple and technically demanding. A single DNA polymerase molecule is immobilized at the bottom of a zero-mode waveguide (ZMW) — a nanophotonic chamber approximately 70 nm in diameter, smaller than the wavelength of visible light. The ZMW creates an observation volume of roughly 20 zeptoliters, confining fluorescence detection to the immediate vicinity of the polymerase active site.
The DNA template is prepared as a SMRTbell — a double-stranded molecule with hairpin adapters ligated to both ends, creating a topologically circular template. When the polymerase begins synthesis, it incorporates fluorescently labeled nucleotides one at a time. Each incorporation event produces a characteristic light pulse as the fluorophore is held in the detection volume for tens of milliseconds before being cleaved and diffusing away. Because each of the four nucleotides carries a distinct fluorophore, the system records nucleotide identity in real time.
The circular SMRTbell structure is what enables HiFi accuracy. After the polymerase completes one pass around the template, it continues around again — and again. A typical HiFi read represents 10–15 passes around the same molecule. The instrument records all passes as a single continuous read; then the CCS algorithm aligns the subreads, identifies the consensus at each position, and outputs a single HiFi read with Q30+ accuracy. Polymerase kinetics — the inter-pulse duration — are also recorded, providing information about base modifications (5mC, 5hmC) that alter polymerase speed.
This direct detection approach avoids the amplification bias inherent to cluster-based sequencing. GC-rich and AT-rich regions are sequenced with the same efficiency as balanced regions, producing uniform coverage across the genome — a critical advantage for whole-genome methylation analysis where bisulfite conversion would otherwise degrade the sample.
We operate all three current-generation PacBio platforms, giving you the flexibility to match the instrument to your project's throughput and budget.
| Platform | SMRT Cells per Run | Typical HiFi Output | Best Suited For |
| Revio | 4 (SMRT Cell 25M) | Up to 360 Gb (90 Gb/cell) | Large eukaryotic genomes, population-scale WGS, Iso-Seq at scale |
| Vega | 1 (SMRT Cell 25M) | Up to 90 Gb | Benchtop HiFi for individual genomes, pilot projects, small labs |
| Sequel IIe | 1 (SMRT Cell 8M) | Up to 30 Gb | Small genomes, targeted amplicons, microbial WGS, legacy projects |
All three platforms use the same core SMRT sequencing chemistry and HiFi read generation pipeline, so data quality is consistent across the range. The choice comes down to throughput, project scale, and turnaround requirements. Revio is our workhorse for large-scale projects — a single SMRT Cell 25M on Revio now delivers up to 120 Gb with SPRQ chemistry, sufficient for 30× HiFi coverage of a human genome. Vega brings the same SMRT Cell 25M technology to a benchtop form factor, ideal for core facilities or labs running one project at a time.
For projects requiring the highest possible read lengths or the ability to run pre-made SMRTbell libraries, our team can optimize run conditions — adjusting loading concentration, extension time, and movie length — to maximize either throughput or read length depending on your project goals.
This is where you find the exact PacBio service for your project. CD Genomics delivers 40+ application-specific PacBio HiFi sequencing services organized by research area. Each linked service page includes detailed workflows, sample requirements, bioinformatics deliverables, and case studies where applicable.
PacBio HiFi reads resolve medically critical genomic regions — HLA loci, pharmacogenes, repeat expansion disorders — that short reads routinely miss or mis-map.
Visit our Human Genomics with Long-Read Sequencing hub for all services.
Large, repetitive, and polyploid genomes — common in plants and many animals — are where PacBio HiFi demonstrates its most dramatic advantage over short reads. A single HiFi read spans repetitive elements that would fragment an Illumina assembly into hundreds of contigs.
Browse all services in our Animal and Plant Genomics with Long-Read Sequencing hub.
HiFi reads enable complete, closed microbial genomes — chromosome(s) plus plasmids — in a single SMRT Cell run. You get a finished genome, not a draft with 50 gaps that need PCR closure.
See all microbial services at our Microbial Genomics with Long-Read Sequencing hub.
Short-read RNA-seq counts fragments, not isoforms. PacBio Iso-Seq reads entire transcripts — 5' end to 3' poly(A) tail — in a single HiFi read, revealing the complete isoform landscape.
Explore all transcriptomics services at our Transcriptomics with Long-Read Sequencing hub.
PacBio SMRT sequencing detects 5mC and 5hmC directly from polymerase kinetics — no bisulfite conversion, no antibody enrichment. The same HiFi read delivers both sequence and methylation status.
Visit our Epigenetics and Methylation Analysis hub.
When you need population-level variation or causal mutation identification, HiFi captures the full variant spectrum — SNVs, indels, SVs — in one experiment.
All resequencing services at Whole-Genome Resequencing.
Visit Pre-Made Library Services and Data Analysis Services.
Every sequencing project starts with a platform question. Below we compare PacBio HiFi with ONT and Illumina across the dimensions that matter for research decision-making.
| Feature | PacBio HiFi (Revio) | Oxford Nanopore (ONT) | Illumina NGS |
| Read Length | 15-25 kb (HiFi) | 10 kb - 4 Mb | 150-300 bp |
| Per-Read Accuracy | Q30+ (>99.9%, CCS) | Q20+ (R10.4.1 duplex) | Q30+ |
| Consensus Mechanism | CCS (10-15 passes) | Duplex basecalling | Bridge amplification |
| Direct RNA Sequencing | No (requires cDNA) | Yes — native RNA | No |
| Modification Detection | Yes — 5mC, 5hmC via kinetics | Yes — 5mC, m6A, Ψ from signal | No (requires bisulfite) |
| GC Bias | Low-Moderate | Minimal | Moderate-High |
| Throughput per Run | Up to 360 Gb (Revio) | Up to 6 Tb (PromethION 48) | Up to 16 Tb (NovaSeq X) |
| Assembly Contiguity | Chromosome-scale (N50 >50 Mb) | Chromosome-scale (N50 >50 Mb) | Fragmented (N50 <1 Mb) |
| Best For | High-accuracy assembly, Iso-Seq, methylation, variant detection in repeats | Ultra-long reads, direct RNA, SV breakpoints, real-time pathogen ID | High-throughput screening, RNA-seq quantification, validated panels |
How to choose: If your primary goal is a high-quality reference genome — one that will serve as the foundation for years of functional studies — PacBio HiFi is the right platform. The Q30+ accuracy means your assembly will be correct at the single-base level, and the 15-25 kb read length means it will be contiguous. If you need the longest possible reads to span extremely large repeats (Mb-scale duplications, centromeric satellites), or if you need direct RNA sequencing with native modification profiling, our Oxford Nanopore Sequencing Technology page covers ONT in detail. If you are running high-throughput screening — counting transcripts per gene or ChIP-seq peaks — Illumina remains cost-effective.
Many of our clients combine platforms strategically: PacBio HiFi for the reference assembly, ONT for structural variant breakpoint mapping, and Illumina for population-scale genotyping. Our project consultation team helps you design the right multi-platform strategy for your research question and budget.
PacBio HiFi sequencing performance depends critically on input DNA quality, particularly molecular weight. HiFi reads average 15-25 kb, and the library preparation process with size selection determines the final read length distribution — which starts with the quality of your extracted DNA.
| Sample Type | Recommended Quantity | Quality Requirement | Critical Notes |
| HMW Genomic DNA | 2-5 µg (standard); 500 ng minimum | OD 260/280: 1.8-2.0; fragment size >30 kb | Avoid vortexing and freeze-thaw cycles; use wide-bore pipette tips; ship on dry ice |
| Total RNA (Iso-Seq) | 1-5 µg | RIN ≥ 7; OD 260/280 ≥ 2.0 | Poly(A) selection included; SMRTbell prepared from full-length cDNA |
| Amplicons (Pooled) | 100-500 ng per pool | Single sharp band on Bioanalyzer | Provide primer sequences and expected amplicon sizes; barcoded multiplexing available |
| Metagenomic DNA | 1-5 µg | OD 260/280: 1.8-2.0; no visible RNA | Include negative extraction controls; avoid bead-beating that shears HMW DNA |
For detailed protocols, shipping instructions, and application-specific requirements, consult our Sample Submission Guideline. Our project managers review every submission and will contact you if additional material or preparation is needed.
PacBio HiFi sequencing produces raw movie data (BAM) that must be processed through CCS to generate HiFi reads. Our standard bioinformatics pipeline processes your data through these stages:
Standard deliverables include: HiFi reads (FASTQ/BAM), genome assembly (FASTA), variant calls (VCF), methylation bedGraph, and a comprehensive QC report with assembly statistics, coverage plots, and BUSCO completeness scores. For advanced analysis — comparative genomics, phylogenomics, pangenome construction, custom visualization — we provide tailored bioinformatics packages.
Visit our PacBio Sequencing Data Analysis page for complete analysis service details.
PacBio HiFi supports genomic DNA, total RNA (for Iso-Seq via cDNA), amplicons, and metagenomic DNA from virtually any organism — human, animal, plant, fungal, bacterial, viral, and environmental samples. The key requirement is sufficient high-molecular-weight input material (>30 kb fragment size for maximum HiFi read length). For challenging samples — FFPE tissue, single cells, low-biomass environmental samples — our team provides custom protocol optimization during project consultation.
PacBio HiFi reads achieve Q30+ accuracy (>99.9%) through CCS (10-15 subread passes). This matches Illumina per-base accuracy and significantly exceeds ONT's single-molecule accuracy (Q20+). For applications where single-base accuracy determines variant calls or assembly quality — genome finishing, rare variant detection, clinical-grade references — HiFi is the preferred platform. ONT offers much longer reads (up to 4 Mb) and native RNA sequencing.
Revio delivers up to 360 Gb across four SMRT Cells 25M — ideal for large eukaryotic genomes and population-scale projects. Vega uses the same SMRT Cell 25M in a benchtop form factor (one cell, up to 90 Gb), perfect for individual genome projects and pilot studies. Sequel IIe uses SMRT Cell 8M (up to 30 Gb), best for bacterial genomes, targeted amplicons, and legacy projects. All three produce identical HiFi data quality — the decision is about throughput and scale.
Yes. During SMRT sequencing, 5mC and 5hmC produce characteristic changes in inter-pulse duration that are captured alongside the sequence. The pb-CpG-tools package analyzes these kinetic signatures to call 5mC at single-nucleotide, single-molecule resolution. One HiFi run provides both genome sequence and complete methylome — no separate library preparation needed.
Bacterial WGS (sample to closed genome) typically completes in 2-3 weeks. Human WGS at 30× HiFi coverage (sample to assembly) usually takes 4-6 weeks. Iso-Seq projects run 3-5 weeks. Timelines include library preparation, sequencing, CCS generation, and standard bioinformatics. Expedited options are available for time-sensitive projects.
We handle the complete workflow — DNA/RNA extraction (if you send tissue or cells), QC, SMRTbell library preparation, sequencing, CCS generation, and bioinformatics. You can also send extracted nucleic acids or prepared SMRTbell libraries. Our PacBio Pre-Made Library Sequencing service accepts finished libraries for data generation only.
Absolutely. HiFi reads are widely used for de novo assembly of non-model organisms precisely because their length spans repetitive regions that fragment short-read assemblies. For transcriptomics, the Iso-Seq pipeline performs reference-free isoform discovery. Our bioinformatics team works routinely with non-model species and creates custom analysis pipelines adapted to your organism's genomic characteristics.
All PacBio services include standard bioinformatics: CCS generation, quality filtering, genome assembly or read alignment, variant calling (for WGS), and QC reporting with BUSCO completeness. Specific deliverables depend on the service — refer to each linked service page for details. Advanced analysis — custom annotation, comparative genomics, phylogenetics, publication figure preparation — is available as an add-on through our PacBio Sequencing Data Analysis service.
1. Genome Assembly Contiguity Comparison — PacBio HiFi assembly (contig N50: 45 Mb, scaffold N50: 128 Mb) vs Illumina assembly (contig N50: 0.5 Mb) for a 1.2 Gb plant genome.
2. HiFi Read Length Distribution — Histogram showing N50 and coverage profile from a standard Revio SMRT Cell 25M run on human genomic DNA.
3. Iso-Seq Transcript Classification — Pie chart: full-splice matches (FSM), incomplete-splice matches (ISM), novel in catalog (NIC), and novel not in catalog (NNC) isoforms from a human brain Iso-Seq project.

References

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