Choose a WES probe panel by checking whether it includes your required regions and produces usable sequence data across those regions under conditions relevant to your study. A larger target size, a longer gene list, or a higher average depth does not, by itself, establish that a panel is the better choice. Comparative exome studies show that target design and coverage uniformity affect how effectively sequencing data cover the regions researchers need.
Start with a defined set of required genes, transcripts, and genomic intervals. Then compare the panel’s design files, region-level sequencing results, sample requirements, and supporting validation data—not just its headline specifications.
What should you define before comparing panels?
Prepare a short project specification before requesting recommendations or quotations:
- Species and reference genome assembly.
- Research objective: germline variant discovery, tumor research, or population-scale analysis.
- Required genes, transcript versions, coding exons, and splice-adjacent regions.
- Any additional targets, such as selected untranslated regions, promoters, or intronic intervals.
- Sample source, available DNA quantity, and DNA quality.
- Variant classes of interest, including single-nucleotide variants, small insertions and deletions, or copy-number changes.
- Sequencing platform, read configuration, sample count, and planned sequencing allocation.
- Intended use: research or a separately validated clinical laboratory workflow.
Make essential targets explicit. “Whole exome” is not a substitute for a coordinate-level specification: commercial panels can differ in their annotation sources, transcript selection, and inclusion of exon-flanking regions. A comparative analysis of exome capture kits demonstrated differences between coverage of core coding regions and coverage of those same regions with added flanking bases.
Does the panel target the regions your study requires?
Request the target BED file—a coordinate-based list of intended genomic intervals—together with its reference assembly, annotation releases, and panel version.
Check the following before comparing target sizes:
- Are the coordinates compatible with your analysis reference?
- Which releases of RefSeq, GENCODE, CCDS, or other annotations were used?
- Are all required coding exons included for the transcripts you intend to analyze?
- What splice-adjacent regions are explicitly targeted?
- Are additional noncoding targets included, excluded, or available as supplementary content?
Where available, request probe-placement or bait intervals separately from the intended target intervals. Probe locations, intended capture targets, and regions used for coverage reporting are not necessarily interchangeable. Confirm which interval set supports each reported percentage.
For a useful design comparison, intersect each panel’s target file with the same project-required interval set. Report missing bases and exons, rather than only counting genes. A gene can appear on a panel list without every relevant exon or transcript being represented.
A larger target footprint is not automatically better. Additional content may be valuable, but it also changes how sequencing reads are distributed. In the comparative study above, target size and uniformity influenced measured coverage after read downsampling; the authors also noted differences in sample preparation and sequencing protocols that limited direct comparisons.
What does “coverage” actually mean?
Separate three questions: whether a region is in the design, whether it receives sufficient usable reads, and whether those reads support reliable analysis.
|
Evaluation measure
|
What it tells you
|
What to confirm
|
|
Design coverage
|
How much of a defined annotation or project interval set overlaps the intended targets
|
Annotation release, reference assembly, denominator, and treatment of flanking bases
|
|
Mean target depth
|
Average sequencing depth across the reporting intervals
|
Whether duplicates and low-quality reads are excluded
|
|
Bases above a depth threshold
|
The fraction of bases meeting a specified coverage level
|
Threshold, interval set, filtering rules, and per-sample results
|
|
Coverage uniformity
|
How evenly sequencing depth is distributed
|
Metric definition and whether comparisons use the same intervals
|
|
On-target rate
|
How much sequencing output is assigned to target regions
|
Whether reads or bases are counted, padding rules, and duplicate handling
|
|
Duplication and library complexity
|
How much sequencing output represents repeated observations of the same library molecules
|
Deduplication method and performance at the intended sequencing allocation
|
|
Callable regions
|
Regions satisfying the analysis pipeline’s requirements for evaluating variants
|
Depth, mapping quality, base quality, and variant-class-specific criteria
|
These distinctions matter because high average depth can coexist with poorly covered exons. Published evaluations assess both design overlap and measured coverage, and demonstrate that filtering ambiguously mapped reads can substantially change the apparent coverage of some regions.
The practical endpoint is callability: whether the resulting data support the analysis you intend to perform. Coverage contributes to callability, but depth alone is not proof of variant-detection sensitivity.
Do not apply one depth threshold to every project. Germline and low-frequency somatic analyses require different supporting evidence. The AMP/CAP oncology panel validation recommendations emphasize establishing assay performance for the intended use and evaluating individual variant types, rather than treating a coverage number as a universal guarantee.
How should you evaluate difficult regions?
Ask suppliers to separate three types of limitation:
- Design gaps: required intervals are absent from the intended target set.
- Capture or library-related undercoverage: targeted intervals receive insufficient usable sequence.
- Mapping limitations: reads are generated but cannot be assigned confidently to the correct locus.
These problems require different responses. Extreme GC content can affect coverage, but it is not the only explanation for difficult exons. Repeats, segmental duplications, and highly homologous genes can create alignment ambiguity. A systematic study of WES coverage bias found that probe design and short-read mappability were major determinants of poorly covered coding regions in the platforms evaluated.
For priority regions, request:
- Exon-level and, where necessary, base-level coverage profiles.
- Results after the mapping-quality and duplicate filters used by your analysis pipeline.
- A list of consistently undercovered or excluded intervals.
- Evidence from more than one sample and from relevant sample types.
- A clear explanation of whether supplementary probes address the observed limitation.
Additional probes may address missing target content, but they do not inherently resolve ambiguous mapping between homologous sequences. If an essential locus remains difficult to analyze, assess an appropriate complementary method rather than assuming that deeper sequencing will solve the problem.
How do study needs change the comparison?
Rare-disease and germline research
Prioritize coverage of the required transcripts, coding exons, and splice-adjacent intervals. Review persistent gaps in genes central to the study, even when the overall exome coverage looks strong.
If copy-number analysis is planned, request evidence for that application and pipeline. Uneven exome coverage can affect both sequence-variant and copy-number analysis; a panel’s suitability for one variant class should not be assumed to establish performance for another.
Cancer research
Define the expected variant allele fractions, sample types, and variant classes before selecting a panel. Low tumor cellularity can increase the depth needed to identify somatic variants, and evidence generated from one specimen type should not automatically be transferred to another.
Ask specifically for performance data relevant to FFPE or other materials you plan to use. If the project includes fusions, microsatellite instability, tumor mutational burden, or other complex outputs, request the target content, analysis method, and supporting evidence for each. Inclusion of associated targets is not equivalent to validation of the complete analysis.
Large cohorts and longitudinal projects
Evaluate consistency across samples, batches, and panel versions—not just the best-performing library. Request documentation of target changes and retain the design files and analysis settings used for each batch.
Comparative studies identify library preparation, sequencing configuration, and analysis pipelines as potential sources of performance differences. For a long-running cohort, include these variables in the comparison alongside the probe panel itself.
Which technical materials should you request?
A supplier evaluation package should make both the design and the reported performance reproducible. Request:
- Versioned target files, reference assembly, and annotation releases.
- Required DNA quantity, quality criteria, and supported sample types.
- Compatible library preparation, capture reagents, adapters, and sequencing configurations.
- Benchmark conditions, including DNA input, insert size, pooling, hybridization protocol, read length, and sequencing allocation.
- Per-sample coverage and uniformity reports, including results for your priority genes.
- Known coverage gaps, mapping limitations, and analysis exclusions.
- Documentation of supplementary target content and panel-version changes.
- Application-specific validation evidence where performance claims extend beyond basic enrichment.
When comparing datasets, use the same project interval set, coverage filters, and sequencing allocation where feasible. Keep sample quality and library preparation as similar as possible. Published benchmarking illustrates the value of downsampling and a shared analysis pipeline, while also showing that these controls do not eliminate every experimental difference.
If shortlisted panels remain difficult to distinguish, run a pilot using representative samples and predefined acceptance criteria.
Project requirements: questions to confirm before ordering
|
Project requirement
|
Question to confirm
|
|
Required genes and transcripts
|
Which required exons or transcript-specific intervals are missing from the design?
|
|
Reference compatibility
|
What genome assembly and annotation releases support the target files?
|
|
Splice-adjacent analysis
|
Which flanking intervals are explicitly targeted, and how are they covered experimentally?
|
|
Essential difficult loci
|
Are limitations caused by missing probes, undercoverage, or ambiguous mapping?
|
|
FFPE, degraded, or low-input DNA
|
What evidence exists under comparable sample and input conditions?
|
|
Germline analysis
|
What fraction of required bases meets the pipeline’s usable-coverage criteria?
|
|
Low-frequency somatic analysis
|
What variant-type-specific sensitivity has been demonstrated at the intended allele fractions?
|
|
Copy-number or other complex analyses
|
What additional workflow and validation requirements apply?
|
|
Large or continuing cohorts
|
How are panel versions, target changes, and batch consistency documented?
|
|
Supplementary targets
|
Can required intervals be added, and how will the expanded design be evaluated?
|
Match the service discussion to a defined project
Dynegene’s whole exome sequencing probes page lists Standard, Tumor, and Heredity versions of its QuarStar Human All Exon Probes 4.0 portfolio, along with RNA-based exome options. It also states that the Standard version supports customized target spike-ins. Treat these as options to investigate against your project specification, not as interchangeable designs or guaranteed outcomes.
Before requesting a quotation, prepare your required intervals, sample details, sequencing plan, and evaluation criteria. Confirm the exact panel version and request documentation relevant to that configuration.
The most useful choice is the panel that includes your essential targets and demonstrates usable coverage under relevant conditions. To discuss a project-specific option, review Dynegene’s WES probe panel options and provide that information for technical review. Product selection alone does not establish suitability for clinical diagnosis; clinical use requires appropriate assay validation and compliance with applicable requirements.