HiFi-C vs CiFi: What’s the Difference in Long-Read 3D Genome Sequencing? (2026)

At a glance:

• Key Takeaways
• What Is HiFi-C Sequencing?
• What Is CiFi Sequencing?
• HiFi-C vs CiFi: Are They the Same Technology?
• How HiFi-C Sequencing Works
• Sample Requirements for HiFi-C Sequencing
• What Data Outputs Does HiFi-C Sequencing Provide?
• When Should Researchers Use HiFi-C Sequencing?
• Side-by-Side Comparison (as of 2026-03-10)
• How to Choose: A Short Decision Path
• FAQ
• References (selected)
• Author

Choosing between closely related methods can stall a project more than picking a platform from scratch. If you're planning chromosome-scale assembly, scaffolding, or phasing and keep hearing both "HiFi-C" and "CiFi," here's the bottom line: they sit in the same long-read 3C/Hi-C paradigm built around PacBio HiFi reads. The practical differences are more about library design, notably enzyme choice, input tolerance, and where the peer-reviewed evidence is strongest.

This article gives you an evidence-led comparison and a pragmatic way to choose, with a focus on assembly/scaffolding outcomes, replicable QC, and sample requirements-so your statement of work and milestones don't drift.

Key Takeaways

• HiFi-C vs CiFi: same long-read 3C/Hi-C paradigm; nuanced differences. CiFi is a named, peer-reviewed long-read multi-contact 3C method; "HiFi-C" is often used for service implementations that pair 3C/Hi-C style libraries with PacBio HiFi sequencing. Peer-reviewed numbers mostly come from CiFi.
• Enzyme choice drives contact density. In CiFi datasets, DpnII yields a median ~17 segments (contacts) per HiFi read vs ~2 for HindIII-higher per-molecule signal helps scaffolding and 3D resolution, while longer HindIII fragments can benefit certain analyses, per the CiFi paper in 2025. See the figures in McGinty et al. - Nature Communications (2025).
• Mapping in repeats and centromeres improves with long reads. In GM12878, CiFi achieved ~83-89% mappable segments in segmental duplications/centromeres vs ~33-37% for short-read Hi-C, according to the peer-reviewed CiFi study in 2025.
• Low-input feasibility is documented. CiFi succeeded around ~62k cells (approximately 370 ng DNA), more than 100 times lower than common bulk Hi-C inputs, per the same paper. Service implementations of HiFi-C often claim similar ranges; confirm your sample specifics in the statement of work.
• Decision rule of thumb: For standard inputs and chromosome-scale scaffolding/phasing, either implementation can work-pick the team with species-proven pipelines and published assembly lifts. For precious low-input samples, published CiFi evidence is stronger. If you need audit-friendly documentation and predictable delivery, consider established HiFi-C service workflows and ask for QC checkpoints up front.

What Is HiFi-C Sequencing?

HiFi-C refers to chromatin conformation capture (3C/Hi-C-style) libraries that are sequenced on PacBio to produce highly accurate HiFi reads preserving multi-contact ligation structures across a single long molecule. Conceptually, it couples a proximity ligation library (crosslink-digest-ligate) with circular consensus sequencing so each HiFi read contains multiple ligated fragments that reflect spatial proximity in the nucleus. The result is a single-molecule, multi-contact record that's much easier to map uniquely across repeats and centromeric DNA than short-read pairs.

What this enables in practice is robust scaffolding of contigs into chromosome-scale assemblies and improved haplotype phasing by exploiting multiple same-molecule contacts at long range. Because the reads are Q30-Q40 and long, at kilobase scale, they are more likely to traverse low-complexity or duplicated sequence and still anchor confidently. For background on 3C/Hi-C and how proximity ligation works, see the internal overview in the LongSeq network's 3C comparison guide at CD Genomics: 3C technologies comparison guide. For a refresher on PacBio HiFi read properties, including Q40 accuracy and CCS passes, see the internal resource: PacBio HiFi sequencing basics.

HiFi-C is not a single peer-reviewed "brand" in the literature the way CiFi is; it's more a family of implementations. That's why most quantitative, peer-reviewed performance claims in the public domain derive from CiFi datasets, while HiFi-C performance tends to be documented by service providers or case-by-case studies.

What Is CiFi Sequencing?

CiFi (Chromatin interaction Fingerprinting) is a peer-reviewed long-read multi-contact 3C method optimized for PacBio HiFi sequencing. It preserves multiple ligated genomic fragments within a single highly accurate read, and its library design explicitly explores enzyme choice trade-offs, such as frequent cutter DpnII versus 6-cutter HindIII. In published results, CiFi demonstrated notable advantages in mapping within repetitive and centromeric regions and workable performance at markedly lower input than conventional bulk Hi-C. See the peer-reviewed report by McGinty and colleagues: the Nature Communications 2025 paper, "CiFi: accurate long-read chromosome conformation capture with low-input requirements."

• On contact density per read and enzyme effects, the paper shows median ~17 segments per read with DpnII, and ~2 with HindIII, with corresponding fragment-length distributions and resolution trade-offs.
• On mapping in repeats and centromeres, CiFi materially outperformed short-read Hi-C in GM12878.
• On input, CiFi libraries succeeded at ~62k cells, approximately 370 ng DNA, in the study design.

For details, consult the peer-reviewed article: according to the authors in the 2025 Nature Communications paper, "CiFi: accurate long-read chromosome conformation capture with low-input requirements," available on PubMed Central as the open-access version: McGinty et al., Nature Communications (2025).

HiFi-C vs CiFi: Are They the Same Technology?

They're best understood as the same long-read 3C/Hi-C paradigm implemented under different labels. CiFi is a formal, peer-reviewed method with a specific library optimization path and published benchmarks. "HiFi-C" is the umbrella term many teams and service providers use for proximity ligation libraries sequenced on PacBio HiFi that yield multi-contact reads. In day-to-day planning, your choice should pivot on practical differences:

• Evidence coverage: CiFi has peer-reviewed, quantitative results for mapping in repeats, contact density by enzyme, and low-input performance in 2025. HiFi-C service implementations may match or approach these metrics but typically publish fewer formal benchmarks; request case studies or data where possible.
• Library parameters and enzyme choices: Both approaches can deploy DpnII or HindIII, and others, trading contact density against fragment size and resolution. Your optimal setting depends on assembly goals and genome characteristics.
• Operational readiness: Established HiFi-C service workflows can de-risk timelines with QC checkpoints, while CiFi's literature gives you confidence around low-input and difficult regions. If your samples are precious, lean toward documented low-input success. If your constraint is delivery and traceability, lean toward mature service pipelines.

HiFi-C and CiFi sequencing both capture chromatin interactions using long-read sequencing.

How HiFi-C Sequencing Works

At a high level, HiFi-C follows a 3C/Hi-C-style proximity ligation workflow but pairs the resulting concatemers with PacBio HiFi sequencing so each read preserves multiple contacts:

1. Crosslinking: Formaldehyde fixation stabilizes in-nucleus DNA-protein-DNA complexes so physically adjacent genomic loci remain tethered during processing.
2. Digestion: A restriction enzyme, commonly DpnII or HindIII, cuts the crosslinked chromatin. Frequent cutters like DpnII increase the number of fragments per molecule; 6-cutters like HindIII yield longer fragments and fewer junctions per read.
3. Proximity ligation: DNA ends within crosslinked complexes are ligated, creating multi-fragment concatemers that encode spatial proximity. Proper ligation conditions and cleanup reduce spurious junctions and chimeras.
4. DNA purification and library prep: After reversing crosslinks, high-molecular-weight DNA is purified and converted into SMRTbell libraries suitable for circular consensus sequencing to obtain HiFi reads.
5. PacBio HiFi sequencing: The polymerase traverses each circularized molecule multiple times to generate a high-accuracy consensus read, Q30-Q40 and often near Q40, that still contains the ligation junctions.
6. Chromatin interaction analysis and scaffolding: Reads are segmented at ligation junctions; each segment is mapped to the reference or draft assembly. Multi-contact information per molecule provides long-range constraints used by scaffolding pipelines, for example hifiasm integration with 3D-DNA or YaHS-style scaffolding. For a primer on long-read chromatin approaches, see the internal overview: Chromatin conformation and long-read methods.

Workflow of HiFi-C sequencing for long-read chromatin interaction mapping.

Sample Requirements for HiFi-C Sequencing

Most projects succeed or fail before sequencing starts-at sample qualification. For HiFi-C or CiFi, three factors matter most: cell or DNA input, DNA integrity, and whether your tissue can be crosslinked and digested reproducibly.

• Input amount: Published CiFi evidence shows successful libraries from about 62k cells, approximately 370 ng DNA. That's unusually low for Hi-C-class methods. Many service workflows describe similar low-input feasibility for HiFi-C, but unless they share species-matched validation data, treat those numbers as indicative rather than guaranteed.
• DNA quality: High-molecular-weight DNA and intact nuclei matter because each molecule ideally carries multiple true ligation products. Degraded input reduces segment count per read and undermines scaffolding value.
• Tissue type and fixation: Solid tissues can work, but fixation and nuclei prep must be tuned so chromatin remains accessible to the chosen enzyme. Over-fixed material can depress ligation efficiency; under-fixed samples can increase noise.
• Enzyme strategy: DpnII usually yields more contacts per read, while HindIII yields fewer but longer fragments. Choose based on the resolution and assembly behavior you need.

For precious samples, request a pilot and define pass-fail checkpoints before the full run. A useful statement of work will specify minimum library yield, target read count, acceptable median segments per read, and mapping thresholds in repetitive regions if those regions are part of your biological question.

For internal service context, you can also review an implementation page here: HiFi-C sequencing service. Use this as a checklist source for sample acceptance criteria, not as a substitute for species-matched validation.

If multiplexing across projects, several modest-size CiFi or HiFi-C libraries may be pooled with HiFi whole-genome sequencing libraries as long as target read counts and per-library index balance are maintained; confirm pooling and demultiplexing plans before sequencing.

When drafting the statement of work, request explicit targets: total HiFi-C or CiFi read count, expected median segments per read if enzyme is predetermined, and minimal usable contacts per gigabase of data. These pre-commitments make downstream "was it enough?" conversations far simpler.

What Data Outputs Does HiFi-C Sequencing Provide?

While deliverables vary by group, you should expect at minimum:

• Raw HiFi reads, FASTQ or BAM, that contain multi-contact concatemers.
• Segment-level mappings and contact pairs derived by splitting reads at ligation junctions, pairs files or equivalent.
• Binned chromatin interaction matrices, .hic or .cool formats, at least at mid-scale resolution suitable for scaffolding and 3D inspection.
• Assembly and scaffolding reports that quantify N50 or NG50 changes and misjoin counts before and after integration with contact data; toolchain parameters used, including mapper, pairtools or juicer, and scaffolder versions, should be recorded for reproducibility.

There's growing evidence that coupling highly accurate HiFi data with proximity ligation improves continuity in tandem repeats and other problematic regions. For background on how long reads assist gap closing in tandem repeats, with short-read Hi-C in that specific work, see the peer-reviewed study by Wen and colleagues: TRFill: synergistic use of HiFi and Hi-C enables gap closing in tandem repeats (2025). Although the scaffolding context differs, the repeat-aware principle carries over when using long-read multi-contact data.

When Should Researchers Use HiFi-C Sequencing?

• Chromosome-scale assembly and scaffolding with phasing: If your goal is to turn high-quality HiFi contigs into chromosome-scale scaffolds with fewer misjoins, long-read 3C implementations, HiFi-C or CiFi, are well-suited. Choose the enzyme or library strategy that aligns with your genome's repeat landscape.
• Structural variation in complex regions: Mapping power in segmental duplications and centromeres is materially better with long-read multi-contact data than with short-read Hi-C in published benchmarks; this translates to cleaner structural variant interpretation near repeats.
• 3D genome organization: While not a ChIP-anchored assay, the multi-contact view supports TAD-scale and loop-scale interpretation; use frequent cutters if you prioritize higher per-molecule contact density.

Also consider the operational route. If you need an end-to-end partner, review a mature service implementation: HiFi-C sequencing service. Keep the tone contractual-request SOP highlights and QC thresholds in the statement of work.

Side-by-Side Comparison (as of 2026-03-10)

Below is a concise, evidence-anchored table. CiFi values cite the peer-reviewed 2025 paper; HiFi-C values reflect typical service descriptions where peer-reviewed numbers aren't published. Short-read Hi-C is shown as a baseline where relevant.

Notes: Values reflect evidence available as of 2026-03-10. Always confirm current chemistries, software versions, and service-level agreements before locking plans.

How to Choose: A Short Decision Path

Start with your primary goal, sample reality, and species evidence.

• If your primary goal is chromosome-scale scaffolding and haplotype phasing with standard inputs, choose either HiFi-C or CiFi. Prefer the implementation with published assembly lifts and misjoin counts for your species or a closely related genome. Ask for a small pilot plus explicit targets, reads, contacts per read, and usable contacts per gigabase, in the statement of work.
• If your samples are precious or ultra-low input, for example single small insects or limited tissues, favor CiFi configurations with documented low-input success in the peer-reviewed literature, then mirror the parameters in your run plan.
• If your genome has extensive repeats or centromeric complexity, prioritize enzyme choices and mapping results validated for your clade. The CiFi paper shows strong mappability in repeats versus short-read Hi-C; a well-tuned HiFi-C service may deliver similar outcomes-request comparable evidence.
• If you operate under tight timelines and compliance scrutiny, prefer established HiFi-C service workflows with named QC checkpoints, version-locked pipelines, and audit-friendly final reports.

Pricing and turnaround disclaimer: Costs, throughput, and turnaround times vary by provider, chemistry, genome size, coverage, and analysis bundle. Treat all planning figures as subject to change and verify them at contract time, as of 2026-03-10.

Also consider a provider option: If you need an end-to-end service with PacBio HiFi and proximity ligation expertise, review the neutral overview here: HiFi-C sequencing service (CD Genomics LongSeq). Use it to frame statement of work questions about inputs, enzyme selection, QC, and deliverables.

FAQ

What does "HiFi-C vs CiFi" really mean for my statement of work language?

Use "long-read 3C/Hi-C with PacBio HiFi, CiFi or HiFi-C," unless your institution requires naming the peer-reviewed method, CiFi. Define the enzyme and target contact density per read to lock expectations.

Can I use fixed tissues instead of cells?

Yes, but expect more upfront optimization. Intact-tissue crosslinking must balance fixation depth against enzyme accessibility. Pilot a small batch and evaluate ligation efficiency and fragment size distributions before the main run.

How many replicates do I need?

Plan for two biological replicates to guard against batch effects and one technical replicate if you're near the low-input threshold. Replicates support matrix concordance checks and more confident scaffolding.

What mapping or assembly metrics should be in my acceptance criteria?

Include: total HiFi-C or CiFi reads, median segments per read or distribution, usable contacts per gigabase, mapping rate within repeats and centromeres, MAPQ at least 1 threshold, and assembly before and after metrics, N50 or NG50 and misjoin counts. Require tool versions and key parameters.

How does this compare to ONT-based Pore-C?

Pore-C is a long-read 3C approach on ONT. If you're comparing long-read ecosystems, review the internal overview here: Pore-C service page. Your choice often turns on read-level accuracy, available instruments, and existing analysis pipelines.

Author

Dr. Yang H., Senior Scientist at CD Genomics - LinkedIn