• Home
• Resource
• Hi-C Sequencing for Plasma Cells and Lymphoid Malignancy Research Samples

Primary plasma cells and lymphoid malignancy research samples are usually not "hard" because Hi-C is exotic. They're hard because they're fragile, heterogeneous, and often limited. Hi-C will still give you a useful genome-wide contact map in this setting, but only if you treat it like an engineering problem: control the input composition (enrichment and purity), control the state you freeze (fixation timing and consistency), and control what you claim (feature-scale realism and RUO interpretation boundaries).

This article is written for PI-led teams and senior bioinformaticians planning Hi-C on enriched hematologic research samples, especially when the downstream goal is not "a pretty heatmap," but defensible, reviewable signals that can be integrated with WGS, RNA-seq, and cytogenetic context.

Key takeaways

• Hi-C is strongest as a genome-wide baseline for compartments and domain-scale organization, and as a way to contextualize large rearrangement-associated contact patterns. Loop-level conclusions are possible, but they are where depth, complexity, and heterogeneity penalties show up first.
• For enriched plasma cells and lymphoid malignancy research samples, the dominant risks are: (1) mixed cell populations diluting the signal, (2) variable or delayed fixation, (3) low complexity libraries where "more sequencing" mostly buys duplicates.
• Treat enrichment as part of the assay design: record the method, purity estimate, viability, clumping, and time-to-fixation. Those metadata often explain more variance than bioinformatics parameters.
• Fresh vs cryopreserved is not a moral preference. It changes failure modes. If you must use cryopreserved input, pilot the workflow and set acceptance criteria before scaling.
• Hi-C-based rearrangement signals are research evidence, not diagnosis. They should be interpreted alongside orthogonal assays and with explicit non-diagnostic language.

What Hi-C can reveal in hematologic research samples

Hi-C is a genome-wide chromosome conformation capture method: you crosslink chromatin contacts, digest DNA, ligate proximal fragments, and sequence the junctions to estimate contact frequency across the genome. The output is a contact map that reflects interaction frequencies in a cell population, not a literal three-dimensional distance measurement.

The foundational Hi-C paper by Lieberman-Aiden and colleagues established the approach as a genome-wide map of long-range contacts and revealed large-scale folding principles in human cells (Lieberman-Aiden et al.'s genome-wide Hi-C map (Science, 2009)). Since then, higher-resolution in situ workflows have made it possible to detect finer structures, including looping principles, when the library and sequencing depth support those claims (Rao et al.'s in situ Hi-C map (Cell, 2014)).

For hematologic research samples, the practical value of Hi-C is that it gives you a single dataset that can support multiple "feature scales" of interpretation.

The feature-scale hierarchy that matters in primary hematologic samples

If you want to avoid scope creep, separate three things you might mean by "3D genome signal":

1. Compartments (A/B): broad segregation patterns that are often detectable at coarser bin sizes.
2. Domains/TADs: mid-scale blocks of enriched local interactions along the diagonal. TADs were formalized in mammalian Hi-C analysis by Dixon et al. (Dixon et al. defined topological domains in mammals (Nature, 2012)).
3. Point interactions/loops: focal peaks that are the most sensitive to library complexity, sequencing depth, and sample heterogeneity.

A conservative way to communicate this hierarchy, especially in enriched primary hematologic samples, is: compartments are usually your first dependable readout; domains can be dependable when the library is strong and the population is not too mixed; loop calls are possible, but they're where you should be most explicit about depth, replicate support, and interpretability.

If your team needs a practical framework for what "resolution" really means in Hi-C analysis and why it depends on valid pairs and library complexity, Imakaev et al.'s guidance remains a useful anchor (Imakaev et al.'s practical Hi-C analysis guidelines ("Hitchhiker's Guide")).

Why hematologic sample composition matters more than most teams expect

Hi-C is population averaged. That sounds obvious, but it's easy to forget when you start interpreting a clean-looking map.

If your enriched fraction is a mixture, you're not just "adding noise." You're mixing contact maps from different cell states. At compartment scale, that can soften or partially cancel A/B patterns. At domain and loop scale, it can reduce peak strength and make boundary calls less stable.

In practice, the shortest path to unusable interpretation isn't a single wet-lab failure. It's a chain of small compromises: modest purity, modest viability, modest clumping, and "we'll fix later." Each one is survivable. Together, they can leave you with a dataset that passes basic QC but cannot support the biological decisions you want.

Sample constraints in primary plasma cells and lymphoid research

Before you plan depth or analysis deliverables, define the constraints that are specific to these samples.

Constraint 1: limited and variable input

Primary plasma cells and lymphoid malignancy research samples often arrive as limited fractions. That pushes you toward low-input workflows or higher PCR amplification. Either way, the limiting factor tends to become library complexity.

Low-input Hi-C methods exist, but they are not a free pass. They are engineering trade-offs that change what failure looks like.

For example, SAFE Hi-C argues that amplification itself can introduce duplicates and bias, and proposes an amplification-free workflow to preserve library complexity (SAFE Hi-C amplification-free library preparation (Genome Biology, 2019)). easy Hi-C proposes a low-input, biotin-free approach designed to reduce DNA loss (easy Hi-C low-input workflow (Cell Reports Methods, 2023)). The common thread is that low-input success is usually about preserving unique ligation products, not just "sequencing harder."

Constraint 2: heterogeneity and purity dilution

Hematologic samples are often mixtures: malignant and non-malignant immune cells, variable tumor fraction, and sometimes dead-cell debris. Hi-C will faithfully report that mixture.

If your downstream question is compartment- or domain-scale reorganization, you can often still learn something from mixed populations, but you should treat the mixture as a design parameter: define what purity you are willing to tolerate for your intended feature scale, and whether you need a matched "background" fraction for context.

Constraint 3: fixation timing and consistency

Fixation is where many primary-sample projects quietly fail.

Over-fixation can reduce restriction enzyme digestion and downstream recovery; under-fixation can fail to preserve interactions through the workflow. Hi-C 2.0 provides practical guidance for handling and fixation, including considerations for suspension cells, which is directly relevant to enriched hematologic suspensions (Hi-C 2.0 optimized protocol (Genome Biology, 2017)). A focused 3C optimization study also shows how fixation duration can trade off with digestion and extraction performance (Fixation-duration trade-offs in 3C workflows (Frontiers in Genetics, 2021)).

The operational point is simple: if you do not standardize "time from final enrichment to fixation," you will spend the rest of the project arguing about biology versus handling.

Constraint 4: fresh vs cryopreserved is a different experiment

You asked for explicit caveats here, and it's worth being blunt.

Fresh and cryopreserved cells can both be used in research workflows, but they do not behave identically. Cryopreservation can change viability, increase debris, and increase clumping risk, which cascades into enrichment yield and fixation consistency.

Key Takeaway: If cryopreserved input is unavoidable, treat it as a separate condition and validate it in a pilot. The goal is not to "make it work" once, but to learn whether it produces stable, interpretable contact maps for the feature scale you care about.

Cell enrichment, purity, and viability before Hi-C

Enrichment is not just a convenience step. In Hi-C, it is a signal-definition step.

What "purity" means in a Hi-C context

In RNA-seq, a mixed population can sometimes be deconvolved computationally. In Hi-C, mixing changes the contact map itself. You cannot "unmix" a contact frequency matrix without strong assumptions.

So the practical definition of purity is: is the enriched fraction composition stable enough that the contact map represents the biology you intend to measure?

Minimum pre-fix QC you should record

For enriched hematologic suspensions, a minimal QC set that meaningfully improves downstream interpretability is:

• viable cell count (and method used)
• viability estimate (dye-based, if available)
• purity estimate (flow cytometry markers, gating summary, or other method)
• clumping/doublet burden (qualitative is better than nothing)
• time from enrichment completion to fixation

These are not bureaucratic checkboxes. They are the variables that explain why two libraries with similar "read counts" produce different interpretability.

If you want a standardized way to think about QC metrics across 3D genomics workflows, CD Genomics summarizes common metrics and how to interpret them in Standardized QC metrics for 3D genomic workflows.

Viability and "dead cell burden" as a structural confounder

Dead cells and debris tend to show up as a practical problem first: clumping, reduced enrichment performance, inconsistent counting, and lower quality downstream nuclei prep.

Even when a library "works," high dead-cell burden can make your contact map harder to interpret because you are no longer sampling a consistent cell population.

A realistic expectation-setting framework

If you want compartments and broad structural patterns, you can sometimes tolerate more variation in purity and viability than if you want reproducible domain boundaries or loop candidates.

So define success in feature-scale terms:

• "We need stable compartment calls across conditions."
• "We need reviewable domain-scale differences."
• "We need loop candidates at specific loci."

Then align pre-fix QC gates and pilot design to that definition.

Handling RoboSep-isolated or immunomagnetically enriched cells

RoboSep and similar immunomagnetic approaches are popular for suspension immune cells because they are fast and scalable. For Hi-C, the method can be compatible, but the handling details matter.

The two failure modes to avoid: delayed fixation and over-fixation

The most common planning mistake is to treat enrichment as a standalone workflow and fixation as a later, separate step.

For chromatin conformation assays, the state you want to measure is the state you crosslink. If enriched cells sit for extended periods before fixation, you are no longer freezing a controlled biological condition.

The second mistake is "fix harder to be safe." Over-crosslinking can reduce restriction enzyme accessibility and digestion efficiency. That trade-off is a known sensitivity point in 3C-derived workflows, and protocol literature stresses that crosslinking conditions must be standardized and tuned to cell type (Hi-C 2.0 optimized protocol (Genome Biology, 2017)).

Bead carryover: what to do, and how strongly to claim it

Peer-reviewed 3C/Hi-C literature is richer on fixation and digestion than on magnetic bead carryover as a specific variable. So in this section, the right tone is: caution and best practice, not a hard "must."

Practical considerations that reduce avoidable variability:

• minimize bead carryover into the fixation and lysis steps when the workflow allows it
• avoid harsh pipetting that increases lysis and debris before fixation
• record whether enrichment was positive- or negative-selection, because this can change what "purity" means and whether beads remain associated with the target fraction

What you should report in methods metadata

If RoboSep or immunomagnetic enrichment is part of the sample story, record:

• enrichment system and kit (or antibody panel)
• positive vs negative selection
• target fraction definition
• purity estimate method
• post-enrichment handling: washes, time-to-fix, temperature

When those metadata are missing, teams often over-attribute differences to "biology."

Detectable 3D genome signals: compartments, domains, and large rearrangement patterns

This is where most proposals quietly over-promise.

Compartments: the most robust first readout

Compartments are a broad-scale signal and are often the most stable feature to interpret when your input is limited or somewhat mixed.

If your study question is "did genome-wide organization shift between conditions," compartments are usually the first place to look. The concept is foundational to early Hi-C maps, including the original Science paper (Lieberman-Aiden et al.'s genome-wide Hi-C map (Science, 2009)).

Domains/TADs: interpretable, but more sensitive to quality

TADs and other domain-scale structures can be very informative, but boundary strength and reproducibility are more sensitive to data quality and handling.

If you see apparent boundary loss or domain weakening in a fragile primary sample, treat it as a hypothesis that must be checked against QC and replicates, not as a conclusion.

A good practice is to treat domain-scale claims as "supported when consistent across replicates and QC passes," and to be explicit about analysis parameters.

Large rearrangement-associated patterns: strong use case, strict language

Hi-C is increasingly used in cancer research to detect or characterize large structural rearrangements and to provide topological context that can be consistent with enhancer hijacking. The key phrase there is "consistent with," not "proves."

NeoLoopFinder provides a peer-reviewed example of using Hi-C to detect enhancer-hijacking events by reconstructing local maps around breakpoints (NeoLoopFinder for enhancer-hijacking detection from Hi-C (Nature Communications, 2021)). EagleC is another example of a method designed to detect a full range of structural variations from Hi-C data at scale (EagleC deep-learning SV detection from Hi-C (Science Advances, 2022)).

For lymphoid contexts specifically, a recent report describes FFPE-compatible Hi-C applied to routine lymphoid cancer biopsies and discusses genome-wide detection of rearrangements and their topological consequences (FFPE-compatible Hi-C for enhancer-hijacking rearrangements in lymphoid biopsies (Cell Genomics, 2026)).

You do not need to turn this into a clinical claim to make it valuable. In RUO language, it supports a reasonable statement: Hi-C can reveal rearrangement-associated contact patterns that help you prioritize breakpoints or loci for orthogonal validation and mechanistic follow-up.

A practical "what you can claim" template

When writing a results summary for these samples, the safest pattern is to tie claims to scale and to controls:

• "We observed compartment-level differences between conditions, consistent across replicates."
• "Domain-scale changes were observed at loci X and Y, but boundary strength was sensitive to library QC; interpretations are limited to RUO context."
• "We observed rearrangement-associated contact blocks suggestive of a large structural event; orthogonal validation is recommended before drawing mechanistic conclusions."

Integrating Hi-C with WGS, RNA-seq, cytogenetics, or other assays

In lymphoid malignancy research samples, you almost always learn more when Hi-C is not asked to be everything.

Integration logic: let each assay do what it is good at

A practical division of labor looks like this:

• WGS: breakpoint-level structural variant calls; copy-number context; clonality estimates.
• RNA-seq: expression consequences; allele-specific or pathway-level shifts.
• Cytogenetics (research context): coarse structural context and orthogonal support for large events.
• Hi-C: topology, compartment/domain context, and breakpoint-adjacent contact patterns that suggest regulatory rewiring.

The best integration questions are not "do these assays agree?" but "what does each one add that the others cannot?"

A concrete example of a mechanistic chain (RUO framing)

• WGS suggests a large event spanning a region.
• Hi-C shows a new interaction block or altered contact pattern consistent with the rearranged topology.
• RNA-seq shows an expression shift in genes within the affected regulatory neighborhood.

None of that is a diagnosis. It is a triangulated research hypothesis with multiple lines of evidence.

When to consider a targeted 3D follow-up

Once a pilot Hi-C dataset points to a small set of loci, a targeted approach can be more efficient than pushing genome-wide depth on every sample.

A useful internal example framework for that escalation logic is CD Genomics' Capture Hi-C variant-to-gene prioritization framework, which illustrates how targeted contacts can support variant-to-gene hypotheses.

If your team is still choosing among assays, the comparative decision guide can help you formalize criteria before you commit: Hi-C vs Micro-C vs Capture Hi-C vs HiChIP.

Pilot study design before cohort expansion

If you only remember one thing: in primary hematologic samples, a pilot is not about proving the biology. It is about proving your workflow and your feature-scale expectations.

Step 1: define the decision the pilot must support

Good pilot decisions are operational:

• "Is enrichment + fixation reproducible enough across samples to support compartment calls?"
• "Do we have enough complexity to see domain-scale features?"
• "Do rearrangement-associated patterns appear in a way that can be cross-checked with WGS/cytogenetics?"

Step 2: choose representative samples, not convenient ones

Pick samples that represent your real constraints:

• a higher-quality enriched sample
• a marginal sample (lower yield, more debris)
• if cryopreserved is in scope, include it in the pilot as its own condition

Step 3: set acceptance criteria before you see the data

If you wait until you see the heatmap to decide what "good" means, you will move the goalposts.

At minimum, define:

• required metadata completeness (purity/viability/time-to-fix)
• minimum number of biological replicates for the comparisons you care about
• what feature scale is required for success (compartments only, compartments + domains, etc.)
• what you will do if complexity is low (repeat wet lab, switch to targeted, reduce claim scope)

If you need a standardized vocabulary for QC, refer back to Standardized QC metrics for 3D genomic workflows so your team is discussing the same things.

Step 4: decide escalation paths early

A pilot should end with a fork, not just a report:

• Proceed to cohort: feature-scale success achieved.
• Repeat / adjust: handling or fixation variability is dominating.
• Re-scope: compartments are stable but domains/loops are not; adjust deliverables.
• Switch assay: if the biological question is loop-first, consider whether Micro-C is justified and feasible. CD Genomics summarizes Micro-C as a higher-resolution option in its Micro-C service, but it should be chosen because the question demands it, not because higher resolution feels inherently better.

Pro Tip: In low-input projects, make "library complexity versus depth" an explicit gate. If a library is duplicate-heavy, buying more sequencing often yields diminishing returns.

RUO limitations and non-diagnostic interpretation boundaries

This entire workflow belongs in research. That changes the language you should use and what you should not imply.

What to state explicitly

• Results are for research use only.
• Hi-C provides evidence about chromatin contact patterns and topology in a sample population.
• Structural patterns may be consistent with rearrangements or regulatory rewiring hypotheses, but they are not diagnostic findings.

What to avoid

• Avoid claims that imply clinical sensitivity/specificity.
• Avoid "detects" phrasing that sounds like a clinical test; prefer "reveals contact patterns consistent with…" or "supports prioritization for orthogonal validation."
• Avoid disease names in this post, per your instruction.

A safe integration sentence you can reuse

Hi-C is often most valuable when it is treated as a topology layer that complements breakpoint- and expression-level assays, rather than as a standalone diagnostic tool.

Submission checklist for hematologic research samples

Below is a submission checklist tuned for enriched suspension-cell Hi-C in hematologic research samples. The goal is to reduce preventable variability and make the final dataset interpretable.

Sample and metadata

• Sample type and source (research context), species, and reference genome build preference.
• Enrichment method: RoboSep/immunomagnetic vs other; positive vs negative selection.
• Target population definition (markers or depletion strategy).
• Purity estimate and method (e.g., flow cytometry markers and gating summary).
• Viability estimate and method.
• Notes on clumping/doublets and any filtration step used.

Handling timeline

• Fresh vs cryopreserved status.
• If cryopreserved: freeze/thaw method, storage duration, and post-thaw recovery conditions.
• Time from final enrichment to fixation.
• Fixation conditions used (formaldehyde concentration/time/temperature) if fixed on site, or whether cells are shipped unfixed for processing.

Quantity and packaging

• Total cell number submitted (post-enrichment).
• Number of aliquots and their approximate counts.
• Buffer composition and shipping conditions (temperature, timeline).

Study design context (so analysis matches intent)

• Comparison groups and batch structure.
• Replicate plan (biological replicates vs technical repeats).
• Primary intended feature scale: compartments only vs compartments + domains vs loop candidates.
• Planned orthogonal assays available for integration (WGS, RNA-seq, cytogenetics) and whether matched samples exist.

Interpretation boundaries

• Confirm RUO, non-diagnostic intent.
• Confirm whether the goal includes hypothesis generation around rearrangements/topology, and what validation path is planned.

---

Author

Dr. Yang H.
Senior Scientist at CD Genomics
LinkedIn: Dr. Yang H. on LinkedIn

Dr. Yang H. supports research teams in selecting fit-for-purpose 3D genomics methods, designing QC-driven workflows for challenging sample types, and producing reviewable, publication-ready bioinformatics for research use only.