[Expert Explanation] How to Prevent Misjudgment in Grandparent DNA Testing - Accuracy Differences and Selection Between STR and SNV Methods

Rewrite date: March 31, 2026

To prevent misjudgment in grandparent DNA testing, it is important to choose an institution using the SNV method (600+ loci), confirm international certifications such as ISO and AABB, and select an institution that discloses its misjudgment track record. The conventional STR method carries a high risk of inconclusive results.

What Is Grandparent DNA Testing? Definition and Uses

Grandparent DNA testing (generational testing) is a test that uses the DNA of the paternal or maternal grandparents to estimate whether a biological blood relationship exists with a grandchild, in cases where the father (or mother) is absent. It is used as an alternative in situations where a direct paternity test is not possible, such as when the father has disappeared, passed away, or refuses to cooperate. (1)

In a standard parent-child DNA test, the DNA of the father (or mother) and child is compared directly, but there are many cases where the parent who should be tested is unable to participate for various reasons. These can include sudden death from a traffic accident or illness, loss of contact after divorce, disappearance while traveling abroad, or refusal to cooperate with testing. In such cases, grandparent DNA testing plays the role of scientifically estimating the blood relationship using the DNA of the grandparents' generation as a clue. (2)

The main purposes of use are as follows.

• Proof of blood relationship for inheritance or paternity claims
• Posthumous paternity acknowledgment (legal procedure support after the father's death)
• Private confirmation of family relationships (testing for peace of mind)
• Official proof for immigration or nationality acquisition
• Confirmation of biological roots in adoption cases

In Japan in particular, the need to prove blood relationships in inheritance disputes has been increasing year by year. In family court applications for posthumous paternity acknowledgment, results of grandparent DNA testing are increasingly being accepted as strong scientific evidence, making testing accuracy more important than ever. In addition, there are cases where immigration authorities require the submission of DNA test results for residency status or nationality acquisition, in which grandparent testing may also be necessary. (3)

The Decisive Difference from Paternity Testing - The Gap in DNA Sharing Rate

Paternity testing and grandparent testing differ fundamentally in the certainty of genetic information. This difference is the biggest factor determining testing accuracy. First, it's important to understand the basic principles of human heredity.

Between parent and child, the child always inherits 50% of DNA from the father and 50% from the mother. This is a biological law with no exceptions. Therefore, in a parent-child DNA test, it is possible to determine with high certainty whether the father's (or mother's) DNA type is present in the child. (3)

On the other hand, the relationship between a grandparent and grandchild involves a more complex genetic mechanism. The amount of DNA a grandchild inherits from a grandparent is theoretically an average of 25%, but this is merely a statistical average. In reality, due to a phenomenon called "chromosomal crossover (recombination)" that occurs during meiosis, the proportion of DNA inherited from a grandparent varies individually, ranging from about 15% to 35%. This range of variation is the fundamental reason grandparent testing is significantly more difficult than paternity testing. (4)

As this table shows, while paternity testing typically achieves a positive probability of 99.99% or higher, grandparent testing shows a significant difference in accuracy depending on the analysis method used. With the conventional STR method, there are cases where the positive probability stays at only 90-99%, and the risk of an "inconclusive (gray zone)" result cannot be ruled out. (4)

Furthermore, in grandparent testing using the STR method, submission of an additional sample from the mother is often required to supplement accuracy. However, since there are situations where the mother is unable to cooperate, whether an institution can achieve high accuracy with just the grandparent and grandchild alone becomes an important criterion when choosing a testing institution. (2)

The Real Accuracy of Grandparent Testing - STR Method vs. SNV Method

3-1. Limitations of the Conventional STR Method

The STR (Short Tandem Repeat) method is the current standard DNA testing technique. It analyzes 24 to 60 DNA regions in accordance with the FBI standard (CODIS system) used in the United States. The STR method is a technology that identifies individuals by detecting differences in the number of repeats of short base sequences present in specific regions of DNA. It provides sufficient accuracy for individual identification in criminal investigations and for standard paternity testing (50% DNA sharing rate). (5)

However, the following structural limitations exist when applying it to grandparent-grandchild relationships (25% DNA sharing rate).

• Because the number of loci analyzed is small, there are cases where blood relationship cannot be statistically confirmed
• Submission of an additional maternal sample is often required to supplement accuracy
• There is a risk of an "inconclusive (gray zone)" result
• In certain ethnic populations, genetic diversity is low, increasing the possibility of misjudgment

The limitations of the STR method stem from the small number of loci analyzed. With a limited number of DNA regions - 24 to 60 - it may not be possible to sufficiently detect the DNA information shared between grandparent and grandchild. In particular, for individuals whose DNA sharing rate deviates significantly from the average of 25% (for example, sharing only about 15%), there is a risk of being judged "inconclusive" even though a blood relationship actually exists. (4)

3-2. Accuracy of the Next-Generation SNV Method (700-Locus Analysis)

The SNV (Single Nucleotide Variant) method is a next-generation technique that extensively analyzes single-nucleotide variations distributed across the entire genome. There are several million or more SNVs (single-nucleotide differences) present across the human genome, and comprehensively examining these makes it possible to build a more detailed genetic profile.

The National Research Institute of Police Science has also reported that examining a large number of SNPs can improve individual identification accuracy, and DDC, the largest private DNA testing company, analyzes more than 2,000 SNP loci. The advantage of the SNV method lies not only in the larger number of loci analyzed but also in the fact that the genetic markers analyzed are distributed evenly across the entire genome, making them less susceptible to the effects of chromosomal recombination. (5)(6)

Because the SNV method analyzes more than 10 times as many loci as the STR method, it achieves overwhelmingly higher accuracy in the statistical likelihood ratio calculation. Submission of an additional maternal sample also becomes unnecessary, and testing can be completed with just 2 people - one grandparent and one grandchild. (2)

Recent research has demonstrated that SNV analysis achieves nearly 100% specificity even in identifying "second-degree" relationships such as grandparent and grandchild. With fully automated sample processing systems, the risk of human error is also minimized. In addition, SNP analysis using next-generation sequencing (NGS) technology is rapidly spreading in the forensic field and is attracting international attention as a method to complement or replace the conventional STR method. (2)(7)(8)

Misjudgment and Trouble Cases and Their Causes

With the spread of DNA testing, the following types of trouble have been reported. Accurately understanding the causes helps avoid harm. Since grandparent DNA testing is technically more difficult than paternity testing, differences in the quality control systems of testing institutions and the analysis methods used directly affect the rate of trouble occurrence.

4-1. Overinterpretation of the "99%" Figure

There have been cases where some institutions reported a statistical value indicating only "a possibility of blood relationship" as a definitive "99% blood relationship confirmed." There are also cases where court-ordered precision testing overturned the result to "no blood relationship."

Cause: With the STR method, which analyzes fewer loci, there are limits to the accuracy of the statistical "likelihood ratio" calculation. The likelihood ratio is the ratio between "the probability of observing the DNA data obtained in the test if a blood relationship exists" and "the probability of observing it if no blood relationship exists." If this ratio is not sufficiently large, the reliability of the determination cannot be guaranteed. The International Society for Forensic Genetics (ISFG) recommends a positive probability of 99.9% or higher (likelihood ratio of 1,000 or more) as the standard for confirming a blood relationship, but grandparent testing using the STR method often fails to meet this standard. (3)

4-2. Sample Contamination

There has been an actual case in which, during a mail-in private test, DNA from a family member living with the grandfather contaminated his sample, resulting in an incorrect positive determination.

Cause: The main cause is human error in sample processing steps that involve manual handling. At institutions that have not introduced fully automated systems, various risks exist, including mislabeling of samples, splash contamination during unsealing, and cross-contamination from shared equipment. There are also cases where contamination occurs at the sample collection stage itself, such as when an oral mucosa swab is taken immediately after eating or drinking, or immediately after contact with another person. (2)

4-3. Misjudgment Due to Similar DNA Types Within a Specific Population

In certain regional or ethnic communities, many people share similar DNA types, so a test that analyzes fewer loci carries a risk of mistakenly identifying an "unrelated person" as a "grandfather."

Cause: If the database used in the STR method does not sufficiently reflect the genetic diversity of a specific population, errors occur in the statistical calculation. This is known as the problem of "population stratification," where the DNA type match rate is naturally higher within genetically close populations, meaning that coincidental matches can occur frequently even among unrelated individuals. The SNV method, which analyzes an order of magnitude more loci, can significantly reduce this risk. (4)(7)(8)

4-4. Misunderstanding Due to Unclear Report Wording

There have also been reported cases where a test report contained only vague expressions such as "a blood relationship is likely" or "a blood relationship cannot be ruled out," leaving the client unable to accurately understand the result. A report that does not include specific figures for the positive probability, negative probability, or likelihood ratio has lower value as legal evidence. A trustworthy institution clearly states the positive probability, negative probability, and likelihood ratio in its report, presenting results in a format that leaves the client with no doubt in making a judgment. (1)

3 Ways to Choose an Institution to Prevent Misjudgment

To prevent misjudgment in grandparent DNA testing, how you choose a testing institution is critically important. Keeping the following three points in mind can help you avoid inaccurate results.

(1) Choose an Institution Using the Next-Generation SNV Method (600+ Loci)

With the STR method, which analyzes only 24 to 50 loci, insufficient information arises for determining blood relationships between grandparent and grandchild. You should choose an institution that uses the SNV method and analyzes at least 600 loci, and ideally 700 or more DNA regions. Since the number of loci analyzed is directly linked to testing accuracy, be sure to check whether the specific number of loci is clearly stated on the testing institution's website or materials. Institutions that do not disclose this figure should be avoided. (2)

(2) Choose an Internationally Accredited Institution (ISO/AABB)

Choosing based solely on "cheap and fast" is risky. It is important to choose an institution with a lab that holds ISO 9001 (quality management) or AABB (American Association of Blood Banks; an international accreditation body for paternity testing) certification. ISO 9001 certification means that a third-party body guarantees that every process, from sample receipt to analysis and reporting, is operated according to international quality management standards. AABB is particularly known as the world's leading accreditation body in the field of DNA paternity and relationship testing. If you plan to submit results as legal evidence, testing at an ISO-accredited institution from the outset is essential. (3)

(3) Choose an Institution That Discloses Its Number of Misjudgment Cases

Many institutions do not disclose the fact that they have made misjudgments. An institution that discloses even unfavorable information is proof that it has confidence in its own quality control. Also be sure to check whether a full refund is guaranteed in the event the test result was incorrect. Transparent institutions publish information such as the number of past cases handled, retest rates, and any history of misjudgment on their website or in their materials. This attitude of disclosure is an important indicator for judging an institution's technical capability and integrity.

Comparison Table of Testing Institutions

Below are the main comparison criteria for choosing a grandparent DNA testing institution. When comparing testing institutions, it is important to judge comprehensively based not only on price but also on technical capability, quality control, and guarantee systems.

In addition to the above, be sure to also check the following items.

• Positive probability in negative cases: High-quality institutions guarantee 0.1% or lower. Low-quality institutions do not specify.
• Need for a maternal sample: Not required at institutions using the SNV method (complete with 2 people only)
• Misjudgment track record: High-quality institutions disclose zero misjudgment cases
• Full refund guarantee: Available at high-quality institutions
• Human error countermeasures: High-quality institutions have introduced fully automated systems

In Japan, seeDNA Genetic Medical Research Institute began full-scale provision of grandparent DNA testing using the SNV method (700-locus analysis) starting in February 2026. It is an institution selected for the Tokyo Metropolitan Government's "Innovative Service Commercialization Support Project" and holds ISO 9001 and Privacy Mark certifications, guaranteeing a positive probability of 99.9% or higher when a grandparent relationship exists, and 0.1% or lower when it does not. (1)(2)

The Specific Process for Requesting a Grandparent DNA Test

For those requesting a grandparent DNA test for the first time, here is an explanation of the general testing process. Understanding the steps in advance helps ensure smooth and accurate results.

1. Selecting and inquiring with a testing institution: Based on the three criteria explained in this article (SNV method, international certification, disclosure of misjudgment history), select a trustworthy institution. Consult in advance by phone or email and communicate the purpose of the test (legal or private use).
2. Applying for and receiving the test kit: The testing institution mails a collection kit that includes an oral mucosa swab. For legal testing, arrangements for a witness should also be made in parallel.
3. Sample collection: Following the instructions included with the kit, collect mucosa from inside the mouth of the grandparent and grandchild respectively. Collect at least 30 minutes after eating or drinking, and seal each sample individually to avoid mixing with others.
4. Returning the sample: Mail the collected sample to the testing institution using the dedicated return envelope. Returning it promptly helps prevent deterioration of DNA quality.
5. DNA analysis at the lab: The testing institution's lab extracts the DNA and analyzes it using the SNV or STR method. At institutions with fully automated systems, the risk of human error is minimized. (2)
6. Receiving the test results: The test report is typically issued within 5 to 10 business days. The report clearly states the positive probability, negative probability, and likelihood ratio, and clients can consult with specialized staff about anything they don't understand.

For legal testing, in addition to the above steps, identity verification and supervision of sample collection by a witness (such as a legal professional like a lawyer) are required. This ensures that the test results meet the requirements to be recognized as legal evidence by courts or administrative bodies. (1)

The Difference Between Legal and Private Testing - Choosing Correctly Based on Purpose

When considering a grandparent DNA test, it is important to clarify in advance whether you should choose "legal testing" or "private testing." While there is no difference in the accuracy of the DNA analysis itself between the two, the sample collection procedures and the legal validity of the results differ significantly.

If there is even a slight possibility of using the results for legal proceedings in the future, we strongly recommend choosing legal testing from the outset. Since results from private testing cannot later be converted into legal evidence, choosing incorrectly would result in double the cost and effort. Also, choosing an institution that supports Apostille certification from the Ministry of Foreign Affairs makes the results usable in overseas court or administrative proceedings as well. (1)

Frequently Asked Questions

Q1. Can a grandparent test be done with just one grandparent and one grandchild?

A. With the conventional STR method, there is a risk of an "inconclusive" result when testing only one grandparent and one grandchild. On the other hand, with an institution using the SNV method (700-locus analysis), a definitive determination with a positive probability of 99.9% or higher is possible with just 2 people - one grandparent and one grandchild. Because the SNV method analyzes more than 10 times as many loci as the STR method, sufficient statistical grounds can be obtained even between a grandparent and grandchild with a DNA sharing rate around 25%. (2)

Q2. Can the results of a grandparent DNA test be used as legal evidence?

A. The procedures differ between "legal testing" and "private testing." To submit results as legal evidence, "legal testing" is required, in which sample collection is conducted after identity verification and in the presence of a legal professional. Also, choosing an institution that supports Apostille certification from the Ministry of Foreign Affairs allows the results to be used in overseas court and administrative procedures as well. If there is any future possibility of legal use, choose legal testing from the outset. (1)

Q3. What is the difference between a "99% paternity index" and "99.9%"?

A. The paternity index (positive probability) is a genetics term indicating the reliability of a test result, not the probability of the parent-child relationship itself. Under international standards (ISFG recommendations), 99.9% or higher is the standard for confirming a blood relationship. While 99% and 99.9% may look like only a 0.9% numerical difference, there is a large difference in the statistical likelihood ratio, with 99.9% being tens of times more reliable. For grandparent testing, it is important to choose an institution that guarantees 99.9% or higher. (3)

Q4. What is a 25% DNA sharing rate? Why is grandparent testing difficult?

A. A child inherits 50% of DNA from each parent. The DNA a grandchild inherits from one grandparent is half of that half, averaging 25%. However, due to chromosomal recombination during meiosis, the actual sharing rate varies individually by about 15% to 35%. This small amount of information combined with the large variation is the main reason determination is difficult with the STR method, which analyzes fewer loci. Because the SNV method analyzes 700 or more loci, it can minimize the impact of this variation and enable accurate determination. (4)(7)

Q5. Does accuracy decrease if I request the test by mail?

A. As long as the oral mucosa swab is properly collected and packaged, accuracy is maintained even after mailing. However, preventing contamination (mixing of another person's DNA) during collection is important. Choosing an institution that uses a fully automated system and eliminates human error in sample processing allows for highly accurate results even with mail-in testing. Avoid eating or drinking for 30 minutes before collection, and seal each sample individually before returning it. (2)

Q6. How much does a grandparent DNA test cost?

A. Costs vary depending on the testing institution and analysis method used, but as a general guide, the STR method costs around 50,000 to 150,000 yen, and the SNV method costs around 100,000 to 200,000 yen. Legal testing tends to be more expensive than private testing due to the addition of witness arrangement and document preparation. It is important to choose an institution based on a comprehensive comparison of accuracy, guarantee systems, and international certification, not just cost.

Q7. Can testing be done if only one grandparent is available?

A. Yes, testing is possible with even just one grandparent. With an institution using the SNV method (700-locus analysis), a positive probability of 99.9% or higher can be achieved with just 2 people - one grandparent and one grandchild. However, with the STR method, it may be necessary to supplement accuracy by gathering samples from both grandparents or by submitting an additional maternal sample. (2)

[References]

(1) seeDNA Genetic Testing & DNA Testing, June 2024
(2) seeDNA Genetic Testing & DNA Testing, November 2025
(3) DNA Profiling, November 2020
(4) DNA Diagnostics Center (DDC), June 2025
(5) National Research Institute of Police Science
(6) PMC, 2025
(7) PR TIMES, March 2026
(8) Mainichi Shimbun, March 2026