Most of the writing about home DNA testing focuses on what the tests can do: 99.99%+ accuracy on paternity trios, reliable kinship confirmation out to second-degree relatives, results in 7 to 10 business days. All true. But there is a short list of situations where no amount of lab skill or marker depth will produce a confident answer, and families landing in one of those situations deserve to know that upfront rather than running three rounds of testing and still having nothing.
This article covers five scenarios where DNA testing hits an inherent limit. For each, the article explains why the limit exists, what the lab can and cannot do about it, and what practical options remain.
Limit 1: Identical twins
Identical twins share essentially the same autosomal DNA because they developed from a single fertilized egg that split after conception. Across the 23 autosomal STR markers a standard paternity test runs, identical twins match at every single one. If one twin is the alleged father of a child and the other twin is the actual biological father, the paternity test cannot distinguish between them. Both will show a 99.99%+ probability of paternity when tested against the child.
This is a fundamental limit of the technology, not a lab quality issue. STR analysis counts repeat units at specific chromosomal locations; identical twins have the same count at every location.
Specialized ultra-deep whole-genome sequencing can sometimes detect the rare post-conception mutations that accumulate differently between twins. This is research-grade testing with a small number of labs offering it, high cost, and longer turnaround. It is not something a standard home or legal paternity test offers. For families facing this scenario, a genetic counselor is the right first stop to understand what deep-sequencing options actually exist and what they cost.
Limit 2: Closely related mother and alleged father
When the mother and alleged father share recent biological lineage, the standard paternity math breaks down in a specific way. Normal paternity math assumes the alleles the child inherited from the mother came only from her side and the alleles that must have come from the father are compared to the alleged father independently. If the mother and alleged father are first cousins, or uncle and niece, or in rare cases siblings, many marker alleles are shared between them before any testing happens.
The result: markers that look like a clear match between child and alleged father may actually reflect alleles the child inherited through the mother's side that happen to match alleles the alleged father carries by shared descent. The reported Combined Paternity Index reads higher than the true biological signal.
Reputable labs flag suspected consanguinity based on the pattern of shared markers across the panel. When flagged, options include running additional autosomal markers (USDC's reflex up to 28 markers helps here), running Y-STR analysis for a paternal-line comparison when applicable, or testing additional relatives to triangulate the biological relationship. Families who already know their situation involves consanguinity should mention it to the lab at order time rather than letting it be discovered later.
For general background on how marker count affects statistical confidence on borderline cases, the guide on what autosomal STR markers are and why paternity tests rely on them walks through the math.
Limit 3: Severely degraded or non-standard samples
Dried cheek swabs collected and stored properly are stable for weeks. Blood samples from a hospital draw are stable in appropriate storage. Standard sample types are not a limit.
The limit comes with non-standard samples: hair samples without roots, old personal effects (toothbrushes stored for years, bandages, cigarette butts), tissue recovered from deceased relatives, decades-old biological material, or samples exposed to heat, moisture, UV, or bacterial degradation for extended periods. Any of these can arrive at the lab with DNA quantity below the threshold the standard panel needs, or with DNA quality low enough that the markers do not amplify reliably.
Labs can sometimes extract usable DNA from degraded samples using specialized techniques (low-copy-number STR analysis, forensic-grade extraction), but these are not standard home test workflows and often require upfront consultation to determine whether the sample is even viable. For most home testing, standard cheek swabs collected per the kit instructions are the right sample type. For non-standard sample questions, calling the lab before ordering is faster than ordering a kit and hoping.
Limit 4: Distant relationships beyond second-degree
Autosomal STR testing is reliable for:
- First-degree relationships: parent-child, full siblings. Signal is strong (50% shared DNA on average), result is confident.
- Second-degree relationships: grandparent-grandchild, half-siblings, aunt or uncle to niece or nephew. Signal is weaker (25% shared on average) but still clearly readable with 23 to 28 markers.
At third-degree and beyond, the biological signal is often too weak for autosomal STRs to resolve confidently:
- Third-degree relationships: first cousins, great-grandparent to great-grandchild, great-aunt or great-uncle. Average shared DNA drops to around 12.5%, which overlaps with the random variation expected in unrelated people. The result is often inconclusive even with reflex markers.
- Fourth-degree and beyond: second cousins, great-great-grandparent. Shared DNA drops below 6%. Standard STR testing cannot confirm these relationships reliably.
For questions about distant relationships, the right tool is typically not a STR panel. Depending on the specific question and the available participants, options include consumer SNP heritage testing (an entirely different technology than paternity testing), Y-STR for strictly paternal-line questions across many generations, or mitochondrial DNA for strictly maternal-line questions. These are different tests, not variations on a paternity test. A conversation with a genetic counselor or a specialized lab can clarify which (if any) will answer a specific distant-relationship question.
Limit 5: Chimeric or mosaic individuals
In rare biological cases, a single person carries two genetically distinct cell populations. This can happen if a twin pregnancy results in one twin absorbing the other's cells during early development (chimerism), or if a mutation during early embryonic development causes different cell populations in different tissues (mosaicism). The person may have one DNA profile in their blood and a different one in their cheek cells, or different profiles between tissues.
For a paternity test, this can mean the DNA extracted from a cheek swab does not match the DNA the person would pass to their children, because the reproductive-tissue DNA is different from the swabbed-tissue DNA. The result reads as a false exclusion when the person is actually the biological parent.
Chimerism and mosaicism are rare, but families with reason to suspect it (prior unusual medical findings, a known absorbed twin pregnancy, documented tissue-specific genetic abnormalities) should raise it with the lab before testing. In some cases, testing a different sample type (blood instead of cheek swab) can resolve a discrepancy. In complex cases, a genetic counselor is the right referral.
What to do when a DNA test cannot give you the answer
Running the same inconclusive test repeatedly does not produce a different result. If a reputable 23 to 28 marker test comes back inconclusive, the productive next steps depend on which limit you are hitting:
- Inconclusive on a kinship test: add a second biological relative to the testing, not re-test the same two people at a different lab. The guide on kinship DNA testing when only one relative is available covers the ranking of which relative to add.
- Suspected consanguinity: run reflex markers, add a lineage test (Y-STR or mitochondrial depending on the line), or add an additional biological relative to triangulate.
- Identical twins as alleged fathers: consult a genetic counselor about ultra-deep sequencing options at research-grade labs.
- Degraded or non-standard sample: consult the lab about alternative sample types or specialized extraction. Do not re-collect the same type of sample from the same source.
- Distant relationships: consider whether a different test type (consumer SNP heritage testing, Y-STR, mitochondrial) can answer the specific question, or whether the question is one that current DNA technology cannot answer.
For most home DNA testing scenarios (paternity, maternity, and standard kinship with reasonably close relatives), these limits do not apply. The 99.99% probability of paternity on a clean trio is a real number backed by real statistics. But when a question lands in one of these edge cases, honest acknowledgment of the limit is more useful than another round of the same test. The National Society of Genetic Counselors can help families find a certified counselor to work through complex scenarios.
0 comments