Close-Kin Mark-Recapture Genotyping: What to Prepare Before Requesting a Project Quote

Figure 1. A quote-ready CKMR project connects biological sampling, reliable kinship genotypes, and a clearly bounded demographic analysis.

Close-kin mark-recapture (CKMR) uses genetically identified relatives as the "recaptures" in a demographic model. A parent-offspring pair or half-sibling pair links sampled individuals through an unobserved parent, and the frequency of such relationships can inform abundance and other population parameters. The laboratory component is substantial, but genotyping alone does not define a CKMR study. Sampling years, age and sex information, geographic coverage, life history, population structure, and the relationship classes used by the model all affect feasibility.

A useful quotation request therefore describes the complete path from field samples to analysis-ready kinship data. It should also distinguish the service provider's genotyping and population-genetic support from the demographic assumptions that require species-specific expertise. Existing SNP Genotyping and GBS capabilities may support different stages, but the method must match the maturity of the marker resource and the scale of the cohort.

CKMR Project Readiness Checklist

1. Species, taxonomic scope, and target population.
2. Primary demographic question and parameter of interest.
3. Number of existing and planned samples by year, site, life stage, and sex.
4. Sampling design, including fisheries-dependent, survey, harvest, biopsy, or non-invasive collection.
5. Age, length, sex, maturity, and other individual-level metadata with uncertainty noted.
6. Expected relationship classes: parent-offspring, half-sibling, full-sibling, or others.
7. Reference genome or assembly status.
8. Existing SNP discovery data, validated markers, and previous genotypes.
9. Sample types, extraction status, DNA quantity, and quality.
10. Expected population structure, migration, mixed stocks, and related populations.
11. Desired laboratory deliverables and downstream kinship or population-genetic analyses.
12. Proposed pilot size, production scale, schedule, and batch constraints.

If several items are unknown, mark them as unknown rather than filling them with assumptions. A feasibility discussion can then identify which unknowns require new data and which can be handled through sensitivity analysis.

Define the Demographic Question First

"Estimate population size" is a starting point, not a complete model objective. CKMR studies can target adult abundance, sex-specific abundance, survival, reproductive output, trend, or other demographic parameters. The relationship class and sampling design must carry information about the chosen parameter. A one-time sample, repeated annual sampling, juvenile-adult design, or harvest-based adult sample creates different comparisons.

Write the intended estimand and time frame:

This definition prevents a mismatch in which the laboratory produces accurate genotypes but the sampling design contains too few informative comparisons for the demographic model.

Parent-Offspring and Half-Sibling Designs Are Not Interchangeable

Figure 2. Relationship classes draw information from different age, cohort, and sampling combinations.

Parent-offspring pairs (POPs) connect an offspring with a sampled parent. Their probability depends on the number of potential breeders and the sampling process. Correct birth or capture timing helps rule out impossible orientations. A juvenile cannot be the parent of an older adult, but age error can make apparently simple filters unreliable.

Half-sibling pairs (HSPs) share one parent. Cross-cohort HSPs can inform abundance and survival when the model accounts for reproductive biology. Full siblings, grandparent-grandchild pairs, avuncular pairs, and other second-degree relatives can overlap with the genomic relatedness expected for half siblings. Relationship classification therefore needs enough independent markers and biological metadata to separate plausible alternatives.

Provide an expected comparison table:

Do not promise that every detected relative pair will enter the model. Pairs can be excluded because their timing is impossible, their class is ambiguous, or their sampling covariates do not fit the model definition.

Describe Samples by Stratum, Not Only as a Total

A request for "3,000 samples" is insufficient. CKMR information comes from comparisons across years, ages, sexes, and locations. A large sample concentrated in one cohort may yield fewer useful comparisons than a smaller, balanced multiyear design. Provide counts in a table with one row per biological and sampling stratum.

Recommended columns include:

1. Sample ID and collection event.
2. Collection year and estimated birth year.
3. Sampling location and method.
4. Life stage, age, and age-estimation method.
5. Length or size with measurement units.
6. Sex and sex-determination method.
7. Maturity or reproductive status where available.
8. Tissue type, preservation method, extraction batch, and DNA metrics.
9. Population or stock label and basis for that label.
10. Availability for repeat extraction or re-genotyping.

Age uncertainty deserves its own field. If age is inferred from length, provide the growth relationship, its source, and measurement error. Recent simulation work shows that misspecified growth curves can bias CKMR abundance estimates. Treat age as measured with uncertainty rather than as an unquestioned integer.

The existing guide to sampling and batch bias in population genomics can help identify confounding between biological strata and laboratory batches. Keep the CKMR design centered on relationship opportunities across cohorts, not only on generic balance.

Match the Genotyping Route to Project Maturity

Figure 3. Discovery, panel validation, and production genotyping require different inputs and acceptance criteria.

No validated SNP resource

If the species lacks reliable markers, begin with discovery using representative individuals across the target population and relevant geographic structure. GBS, RAD-style approaches, or whole-genome sequencing may be considered depending on reference quality and project goals. Discovery samples should represent the diversity expected in production. They should not consist only of close relatives or one locality.

Candidate SNPs exist but no production panel is validated

The next step is filtering for genotype quality, allele frequency, genome distribution, LD, sequence uniqueness, and assay feasibility. CKMR usually benefits from many independent markers because it must distinguish first- and second-degree relationships from unrelated or other related pairs. A Linkage Disequilibrium Analysis can help avoid overweighting correlated loci.

A validated panel already exists

Provide the marker manifest, assay chemistry, primer or probe version, reference assembly, allele coding, prior call rate, replicate concordance, missingness, and populations in which it was validated. A panel developed for population assignment or parentage may be useful, but CKMR relationship classification can demand different marker density and error characterization. Transferability should be tested rather than assumed.

If field samples are degraded or low input, include those conditions in the pilot. The resource on non-invasive wildlife genotyping outlines how allelic dropout and false alleles can affect relationship inference.

Marker Properties Needed for Kinship Classification

CKMR marker design should support accurate relatedness likelihoods across the relevant populations. Useful properties include high call rate, low replicate error, broad genome distribution, adequate MAF, limited LD, consistent allele coding, and stable performance across sample types and batches. Markers linked to sex determination may also support demographic stratification when field sex is unavailable, but they should be validated separately from the autosomal kinship set.

Marker count alone is not an acceptance criterion. Evaluate classification distributions for POPs, HSPs, full siblings, other second-degree relatives, and unrelated pairs using simulations and known relationships where available. Include the measured genotyping error and missingness rather than idealized zero-error assumptions.

Population Structure Must Enter Before Kinship Calling

Population structure can alter background relatedness and allele frequencies. Mixed stocks or spatially structured populations may increase false relatedness signals when a single pooled frequency model is used. Conversely, aggressive filtering by location can remove genuine relatives that disperse between sampling areas.

Describe known stocks, migration, spawning areas, subpopulations, and admixture. Use exploratory PCA analysis and population structure analysis where appropriate. The demographic and genetics teams should then decide whether to stratify frequency estimates, model population membership, exclude migrants, or incorporate movement explicitly.

Do not use PCA clusters as a substitute for a biological population definition. Cluster interpretation depends on sampling and marker choice, and close relatives can influence the apparent structure.

Design the Pilot Around Failure Modes

A pilot should test both laboratory performance and kinship separability. Include samples across years, sites, DNA qualities, and expected populations, plus replicates and known relationships if available. If production will span many plates or runs, place bridge samples across batches.

Predefine pilot outputs:

1. DNA and library QC by sample type.
2. Sample call rate and reasons for failure.
3. Locus call rate, depth, allele balance, and batch behavior.
4. Replicate concordance and sample identity checks.
5. Missingness by year, site, sex, age, tissue, and extraction batch.
6. LD-pruned marker count and allele frequencies by population.
7. Simulated and empirical classification of relationship classes.
8. Expected production rerun and dropout rates.
9. Versioned manifest and recommended production method.

The pilot should also estimate how many samples produce usable genotypes. Pricing based only on received samples can be misleading when difficult material has a substantial repeat or failure rate. The DNA sample suitability guide can support this pre-submission review.

Separate Genotyping Support from Demographic Assumptions

The provider can generate and QC genotypes, estimate allele frequencies, assess LD and structure, identify candidate kin under a specified framework, and deliver analysis-ready files. A demographic CKMR model additionally requires choices about survival, breeding probability, reproductive skew, fecundity, migration, population closure, sampling selectivity, and age uncertainty. These assumptions must reflect the species and sampling process.

Use a responsibility table in the quotation request:

This division does not prevent an integrated project. It makes the quote clearer by showing which model inputs already exist and which require development.

What Drives Feasibility and Price?

The major drivers are not limited to sample count. They include whether SNP discovery is required, panel size and assay maturity, DNA extraction needs, sample quality, number of plates and batches, controls and replicates, rerun assumptions, reference-genome quality, population-structure analysis, kinship classification scope, metadata cleaning, and demographic modelling responsibilities.

For an informative quote, provide three numbers:

1. Samples already collected and ready for extraction.
2. Additional samples expected by year and stratum.
3. Samples intended for pilot versus production.

Also state whether raw reads, genotype calls, VCF, marker QC, sample QC, kinship likelihoods, relationship calls, or demographic model outputs are required. Use the general population genomics quote checklist for administrative details, while keeping the scientific brief CKMR-specific.

Minimum Package for a Feasibility Request

Submit:

1. A one-page statement of the demographic question and population unit.
2. A sample-by-stratum count table and individual metadata dictionary.
3. Reference assembly and existing SNP-resource summary.
4. Sample type, preservation, extraction, quantity, and quality information.
5. Expected kin classes and comparison groups across years.
6. Known population structure and movement information.
7. Proposed pilot and production scale.
8. Required laboratory, bioinformatic, kinship, and modelling deliverables.
9. Schedule, sample arrival pattern, and decision deadlines.

If individual metadata cannot be shared initially, provide an anonymized schema and aggregated counts. The goal is to expose design constraints before a price is tied to an incomplete scope.

What a Kin Match Does Not Establish by Itself

A high-scoring candidate relationship is evidence under a specified marker set, population model, and error model; it is not a demographic conclusion by itself. A CKMR analysis must still account for age, sex, sampling year, location, maturity, population structure, and the probability that particular relatives were available to be sampled. Misclassified relationship types, unmodeled substructure, or duplicated individuals can propagate into abundance estimates even when genotypes are technically sound.

Troubleshoot the pilot by separating laboratory, marker, and demographic-model failures. Sample-wide low call quality points toward input or processing; recurrent locus failure suggests assay design or mapping; an excess of apparent relatives in one sampling stratum may indicate structure, close sampling of a family group, or metadata error. Confirm identities, examine replicate concordance, test alternative relationship classes, and review whether the reference allele frequencies match each population component before expanding the cohort.

Published CKMR studies demonstrate feasibility in particular species and sampling programs, but their SNP counts, precision, and sampling ratios should not be transferred as universal requirements. A panel validated for parentage may add value as a candidate set, yet it may not separate half siblings from unrelated pairs under the intended design. Independent field data, simulation under project-specific assumptions, sensitivity analysis, and additional genotyping validation are needed before demographic deployment. Our support covers research-use genotyping and related bioinformatics; demographic interpretation remains dependent on the investigator's sampling design and model assumptions.

Frequently Asked Questions

For research purposes only. The information and services described here are not intended for clinical diagnosis, therapeutic decisions, or personal health assessment.

  Related Services

Sequencing

Whole Exome Sequencing for Population Genetics

Bioinformatics

Genetic Diversity Analysis

Genome-wide Association Analysis Service

QTL Location Analysis Service

Whole Genome Re-sequencing for Population Genetics

Genotyping By Sequencing (GBS) Service

  Download

Power Your Regulatory Discoveries with DAP-seq: A MYB-GLK Case Study

EM-seq Services for High-Resolution DNA Methylation Analysis

Multi-Omics Strategies to Decode Molecular Mechanisms at Every Level

Explore Customized Bioinformatics Services: Case Studies & Insights