Disclaimer: CD Genomics provides sequencing and bioinformatics services for research use only, not for clinical or personal diagnosis. This article is for informational purposes only and does not constitute medical advice.

Figure 1: cfDNA fragmentomics reads genome-wide fragmentation patterns from low-pass WGS, while targeted cancer panels focus on mutations within predefined gene sets. Each approach answers different research questions.

If your liquid biopsy study needs to detect cancer signals, you face a choice: a targeted panel that hunts known mutations in a few hundred genes, or low-pass WGS fragmentomics that reads genome-wide chromatin fragmentation patterns without looking for any specific mutation. This guide compares what each approach reveals, where each one wins, and when using both together produces the strongest evidence.

CD Genomics provides cfDNA fragmentomics (low-pass WGS), cancer panel sequencing, and targeted region sequencing services.

TL;DR

• Targeted panels win when you know which mutations matter — they achieve high sensitivity for specific variants, work with established clinical workflows, and cost moderately per sample.
• Fragmentomics wins when no trackable mutation exists, when tissue-of-origin matters, or when the research question spans the whole genome — it detects cancer signals in tumors with low mutation burden and reveals chromatin-level biology that panels cannot access.
• The 2025 consensus: fragmentomics features can now be extracted from targeted panel data, and combining both modalities in a single assay improves sensitivity beyond either approach alone.
• What to do next: define your research question first — are you hunting mutations, detecting presence of cancer, or both? Your answer determines the method.

Two Ways to Read cfDNA: Fragment Patterns vs Known Mutations

Targeted panels and fragmentomics ask fundamentally different questions of the same blood sample. Understanding the difference avoids the most common mistake in liquid biopsy study design — choosing a method before defining the question.

What targeted panels measure

A targeted cancer panel sequences a preselected set of genes — typically 50 to 500 — at high depth, often several thousand-fold coverage, to detect single-nucleotide variants, small insertions and deletions, and copy number changes at specific loci. The output is a list of detected mutations with variant allele frequencies. This approach is powerful when the relevant mutations are known — for example, EGFR L858R in lung cancer or KRAS G12D in pancreatic cancer. At CD Genomics, cancer panel sequencing covers clinically and biologically relevant gene sets at the depth needed for reliable variant calling in plasma cfDNA.

What fragmentomics measures

Fragmentomics uses low-pass whole-genome sequencing — typically 0.1× to 6× coverage — to analyze how cfDNA breaks apart across the entire genome. It does not look for individual mutations. Instead, it measures fragment size distributions, end motif frequencies, and nucleosome positioning patterns that collectively reveal whether tumor-derived DNA is present and, often, which tissue it came from (Cristiano et al., 2019). The CD Genomics cfDNA fragmentomics service applies these genome-wide analyses at depths matched to study goals, from screening to comprehensive biomarker discovery.

Why the distinction matters for study design

Choosing between them is not about which technology is better — it is about which biological signal answers your research question. If your question is "does this patient's tumor carry an EGFR T790M resistance mutation?", only a targeted panel (or equivalent deep sequencing of that locus) can answer it. If your question is "does this individual have cancer, and if so, where might it be?", fragmentomics provides information that mutation-only approaches cannot. Many studies need both answers, which is why the field is moving toward integrated approaches (Nguyen et al., 2025).

What Targeted Cancer Panels Detect Best

Targeted panels excel at questions that have a specific molecular target. When the biology is well-characterized and the relevant genes are known, deep targeted sequencing remains the most efficient route to a quantitative answer.

Known mutations at high sensitivity

Panels routinely achieve variant allele frequency detection limits below 0.1% for hotspot mutations, enabled by sequencing depths of several thousand-fold at targeted loci. This makes panels the method of choice for tracking known resistance mutations, monitoring minimal residual disease with a pre-identified mutation signature, and screening for actionable variants in genes with established clinical relevance. Whole exome sequencing extends this principle to all coding regions when a broader mutational survey is needed.

Established analysis workflows and benchmarks

Targeted panel bioinformatics pipelines are mature and standardized. Variant calling, filtering, and annotation follow well-validated workflows with published performance benchmarks. For CRO-managed studies and multi-site trials, this standardization reduces analytical variability and simplifies cross-site data harmonization. The trade-off is that panels only see what they are designed to see — a 500-gene panel covers less than 2% of the genome.

When a panel is sufficient for your research goal

A targeted panel is the right primary tool when your study addresses a specific gene or pathway hypothesis, when the cohort is selected for a known tumor type with well-characterized mutational drivers, or when you need quantitative variant-level data for correlative analysis with clinical endpoints. In these scenarios, adding fragmentomics may provide supplementary biological context but does not change the primary analytical strategy.

What Fragmentomics Adds That Panels Miss

Fragmentomics detects cancer signals that are invisible to mutation-based panels. This is not a marginal advantage — it fundamentally changes which cancers can be detected and which biological questions can be asked.

Detecting cancer without a known mutation

Many cancers — particularly those with low tumor mutational burden, including some breast cancers, sarcomas, and pediatric tumors — carry few or no detectable mutations in standard panel genes. Fragmentomics detects these cancers through chromatin-level alterations rather than sequence-level changes. The DELFI-TF approach detected tumor DNA in samples where mutations were undetectable by ddPCR, and baseline fragmentomics scores independently predicted overall survival (van 't Erve et al., 2024).

Tissue-of-origin information without a tissue biopsy

Because cfDNA fragmentation patterns reflect the chromatin organization of the cells of origin — including tissue-specific nucleosome positioning and transcription factor binding — fragmentomics can infer the tissue source of tumor-derived cfDNA. The DELFI approach correctly identified the tissue of origin in approximately 75% of cases (Cristiano et al., 2019). This capability is especially valuable in multi-cancer early detection studies and in cancers of unknown primary, where a targeted panel can report mutations but cannot say where they came from.

Genome-wide coverage at lower sequencing cost

At 0.5–6× coverage, low-pass WGS surveys the entire genome. While the per-base resolution is far lower than a targeted panel, the breadth of coverage enables analyses — copy number alterations across all chromosomes, genome-wide fragmentation profiles, and chromatin accessibility inference at any regulatory region — that are geometrically impossible from a panel covering 0.1–2% of the genome. The per-sample sequencing cost of low-pass WGS at these depths is generally lower than that of a high-depth targeted panel, though library preparation costs are comparable.

Side-by-Side Comparison

When to Choose Fragmentomics Over a Targeted Panel

Fragmentomics becomes the preferred approach when the research question is broader than any single gene set. These are the scenarios where fragmentomics adds the most value.

Multi-cancer early detection and screening studies

When the goal is to detect the presence of cancer across multiple tumor types in an asymptomatic or high-risk population, fragmentomics provides the breadth needed. The GUIDE study, which combined methylation and fragmentomics on a targeted panel of 1,656 markers, achieved 82.8% sensitivity at 95.8% specificity across five gastrointestinal cancers, with 81.5% sensitivity in an independent cohort where two-thirds of cases were stage I/II (Huang et al., 2025). For studies that aim to detect cancer before it is clinically apparent, genome-wide signals consistently outperform mutation-only approaches because early-stage tumors often lack detectable mutations in any given panel gene.

Cancers with low tumor mutational burden

Breast cancer, sarcomas, and many pediatric cancers carry few somatic mutations. In the SPOT-MAS Plus study, hotspot mutations were detected in 96.5% of liver cancer cases but in only a minority of breast and gastric cancers — while the combined methylation-plus-fragmentomics signal detected cancers across all five types (Nguyen et al., 2025). If your study cohort includes tumor types with low TMB or if the TMB distribution is unknown, fragmentomics provides a detection signal that does not depend on mutation burden.

When tissue-of-origin identification changes the study outcome

In cancers of unknown primary, in multi-cancer cohorts where the tissue source matters for downstream analysis, or in studies evaluating whether a detected signal represents a primary tumor or a metastasis, fragmentomics-based tissue-of-origin inference provides actionable information that targeted panels cannot. For detailed guidance on combining fragmentomics with other liquid biopsy signals, see our article on multi-signal liquid biopsy strategies.

When a Targeted Panel Is the Better Choice

Targeted panels remain the right tool when the analytical goal is molecular precision at specific genomic positions, rather than genome-wide pattern detection.

Tracking known resistance mutations during treatment

When a patient's tumor carries a characterized driver mutation — EGFR, KRAS, BRAF, or similar — and the research question is whether a resistance mutation emerges during therapy, a targeted panel provides quantitative variant allele frequencies at the specific positions that matter. Fragmentomics cannot resolve individual mutations at the variant level and is not suitable for this task.

Studies requiring variant-level statistical power

If your study design requires correlating specific mutations with outcomes — for example, testing whether TP53 mutation status associates with treatment response — only a panel (or equivalent deep sequencing) provides the per-variant quantitative data needed. Fragmentomics provides genome-wide features that can be correlated with outcomes, but these features are aggregate measurements rather than variant-level readouts.

When existing panel data can be re-analyzed

A major 2025 finding from Helzer and colleagues demonstrated that fragmentomics features can be extracted from existing targeted panel sequencing data, with normalized depth across all captured exons achieving an AUROC of 0.943 for cancer detection (Helzer et al., 2025). This means that if you already have targeted panel data, you may be able to extract fragmentomics insights without additional sequencing. However, dedicated low-pass WGS still provides richer fragmentomics features than panel-derived metrics, and panel-based fragmentomics is best viewed as a supplementary analysis rather than a replacement for dedicated fragmentomics sequencing.

Using Both: The Complementary Strategy

The strongest signal often comes from combining both approaches. The 2024–2025 literature consistently shows that multimodal integration outperforms any single modality.

Figure 2: A complementary strategy combining low-pass WGS fragmentomics (genome-wide signals, tissue-of-origin) with targeted panel sequencing (high-sensitivity mutation detection) from the same plasma sample.

How integrated assays work in practice

Integrated approaches extract multiple data types from a single blood draw. SPOT-MAS Plus combined methylation, fragmentomics, and 700 hotspot mutations in a single targeted amplicon workflow, improving overall sensitivity to 78.5% compared with the methylation-plus-fragmentomics baseline (Nguyen et al., 2025). The GUIDE study's GutSeer assay integrated targeted methylation at 1,656 markers with fragmentomics features through a deep neural network, and the combined model significantly outperformed WGS-based fragmentomics alone (AUC 0.963 vs 0.887) (Huang et al., 2025).

What a combined workflow looks like for your project

A practical combined approach uses low-pass WGS at 1–6× for genome-wide fragmentomics and copy number analysis, plus a targeted panel — either a commercial panel or a custom design through targeted region sequencing — for high-sensitivity mutation detection at genes relevant to your study. The two datasets are generated from aliquots of the same cfDNA library, avoiding the need for separate sample preparations. Analysis integrates mutation calls, fragmentomics features, and (if methylation data is also collected) methylation markers into a single classifier or correlative analysis.

When integration is worth the additional cost and complexity

Integration adds value when any of the following is true: the cohort includes cancer types with variable TMB; tissue-of-origin matters for the study endpoint; false negatives from either method alone would compromise the study; or the project aims to discover biomarkers that combine genetic and epigenetic signals. In these cases, the incremental cost of adding fragmentomics to a panel-based study — or vice versa — is typically justified by the gain in detection sensitivity and biological resolution.

Cost Considerations: Fragmentomics vs Panels vs Both

Per-sample cost comparisons are directional only, but understanding cost drivers helps in scoping discussions with providers.

What drives cost in each approach

For low-pass WGS fragmentomics, the dominant cost driver is sequencing depth — 0.5× is cheaper than 6×, and the choice depends on which fragmentomics features the study requires. Library preparation costs are constant per sample. For targeted panels, the dominant drivers are panel size (number of genes), sequencing depth per target, and whether the panel is commercial (off-the-shelf) or custom-designed. For combined workflows, the cost is roughly additive, though shared library preparation reduces some duplication.

Directional cost tiers

As a rough order of magnitude: ultra-low-pass WGS at 0.1–0.5× typically falls into the lowest per-sample sequencing cost tier; low-pass WGS at 1–6× is comparable to or slightly less expensive than a moderate-sized targeted panel; high-depth targeted panels and whole exome sequencing occupy the middle tier; and combined fragmentomics-plus-panel workflows occupy the upper tier. These are directional comparisons only — actual costs depend on sample volume, study design, and the specific deliverables included in the scope of work.

Decision Framework: Which Approach Fits Your Study?

Use this decision framework to align your study goals with the appropriate liquid biopsy approach before contacting a provider.

1. Is your primary goal to detect a specific mutation or track a known variant? → Targeted panel.
2. Are you screening for cancer across multiple types without a known driver mutation? → Fragmentomics, or fragmentomics plus a panel.
3. Does tissue-of-origin identification matter for your endpoint? → Fragmentomics (targeted panels cannot provide this).
4. Is your cohort enriched for low-TMB cancers or is TMB unknown? → Fragmentomics provides mutation-independent detection; panels may miss low-TMB tumors.
5. Do you already have targeted panel data from this cohort? → Consider extracting fragmentomics features from existing data before ordering additional sequencing.
6. Does your study require variant-level quantitative data for correlative analysis? → Targeted panel or whole exome sequencing.
7. Is your study primarily a biomarker discovery effort with unknown targets? → Fragmentomics, potentially combined with a panel for maximal feature space.

If multiple answers point in different directions, a combined approach may be the most robust. For guidance on study design parameters — sample requirements, depth selection, and control strategies — see our companion article on cfDNA fragmentomics study design. For understanding what deliverables to expect, see the cfDNA fragmentomics report interpretation guide.

Figure 3: A decision tree for choosing between fragmentomics, targeted panels, or a combined approach based on research question, cohort characteristics, and study endpoints.

FAQ

Can fragmentomics replace targeted panels entirely?

No. Fragmentomics and targeted panels answer different questions. Fragmentomics detects genome-wide chromatin-level alterations and can infer the presence of cancer without a known mutation, but it cannot resolve individual sequence variants at the sensitivity needed for resistance mutation tracking or minimal residual disease monitoring with a known signature. Targeted panels remain the method of choice when specific mutations must be quantified. The two approaches are complementary, and many studies benefit from both.

If I already have targeted panel data, can I extract fragmentomics information from it?

Partially. Helzer et al. (2025) demonstrated that fragmentomics features — particularly normalized read depth across all captured exons — can be extracted from existing targeted panel data with an AUROC of 0.943 for cancer detection. However, panel-derived fragmentomics features are less rich than those from dedicated low-pass WGS, which provides genome-wide coverage. Panel-derived fragmentomics is best viewed as a supplementary analysis rather than a full replacement for WGS-based fragmentomics when comprehensive fragmentomics features are needed.

What is the minimum ctDNA fraction each method can detect?

Targeted panels can detect individual mutations at variant allele fractions below 0.1% at the sequenced loci, enabled by deep coverage at targeted positions. Fragmentomics detection limits depend on which features are analyzed: tumor fraction estimation from copy number alterations is reliable down to approximately 3% at 0.1–1× depth; fragmentation pattern analysis with machine learning can detect cancer signals at ctDNA fractions below 1% at 5–6× depth; and multimodal approaches combining fragmentomics with methylation have demonstrated detection at ctDNA fractions as low as 10⁻⁵ in experimental settings.

Is a combined fragmentomics-plus-panel approach always better?

Not always. Combined approaches add cost and analytical complexity, and the incremental benefit depends on the study context. If your cohort consists exclusively of tumors with well-characterized, high-frequency driver mutations, adding fragmentomics may provide only marginal additional information. If your cohort includes multiple cancer types with variable TMB, or if tissue-of-origin identification matters, the combined approach delivers substantially more information. Evaluate the incremental gain against the incremental cost for your specific study design.

How do I choose between a commercial panel and custom targeted sequencing?

Commercial panels (e.g., FoundationOne Liquid CDx, Guardant360) offer standardized, validated workflows with established performance benchmarks and are suitable for studies where the gene content aligns with published panels. Custom targeted sequencing through targeted region sequencing is preferable when your study requires specific genomic regions not covered by commercial panels, when you need to add content to an existing panel design, or when cost optimization for large cohorts is a priority. Custom panels require an initial design and validation investment but can be more cost-effective at scale.

Research Use Only. This article is for educational and informational purposes and does not constitute medical or diagnostic advice.