[Expert Supervised] Is a 99.99% Paternity Probability Reliable?

Rewritten on: September 2, 2025

This article explains the scientific basis and reliability of a 99.99% probability of paternity in DNA paternity testing. Our experts cover in detail how the probability is calculated, the precision difference between 99% and 99.99%, why the result never reaches 100%, and how to choose a reliable testing laboratory.

Introduction

For anyone considering DNA paternity testing, one of the biggest questions is just how reliable a "99.99% probability of paternity" really is. To answer this up front: a 99.99% probability of paternity is an extremely reliable figure both scientifically and legally, and in practice it is treated as "de facto proof" [ref:1].

This figure means that the chance of another man's DNA matching purely by coincidence is less than 0.01% — statistically speaking, this qualifies as "conclusive evidence." That said, the result never reaches 100% in theory, and the reliability of the test results also depends heavily on the quality management system of the testing laboratory, so choosing the right laboratory requires careful attention.

This article provides a detailed, expert explanation of everything from the exact meaning and calculation method of the probability of paternity, the practical difference brought about by the precision gap between 99% and 99.99%, the scientific reason the figure never reaches 100%, to concrete points for choosing a trustworthy testing laboratory.

What Is the Probability of Paternity?

The Probability of Paternity (PP) is a numerical value indicating the likelihood that a given man is the biological father of a child. This probability is one of the most important indicators in DNA testing, used to objectively evaluate the reliability of test results.

For example, when a figure of "99.99%" is given, it means that "the tested man is more than 9,999 times more likely to be the father than an unrelated man chosen at random." Stated a little more precisely, it indicates that "even if there happened to be another man within the same population group who shared the same genetic profile as the tested man (alleged father), the alleged father who took part in the test is overwhelmingly more likely to be the biological father" [ref:2].

When calculating the probability of paternity, a baseline value called the "Prior Probability" is set. Typically, the prior probability is assumed to be 0.5 (50%). This is based on the neutral premise that, before testing, the likelihood of "being the father" and "not being the father" are considered equal. This prior probability is combined with the Likelihood Ratio obtained from DNA analysis to calculate the final probability of paternity.

Why Can 99.99% Be Considered "Reliable"?

The reason a 99.99% probability of paternity is considered "reliable" can be explained from both a scientific and an international-standards perspective.

• The minimum standard for probability of paternity required by the AABB (American Association of Blood Banks) is 99.0%, but many reputable testing laboratories guarantee precision of 99.99% or higher, far exceeding this minimum [ref:3].
• DNA paternity testing does not rely on a single genetic locus; it simultaneously analyzes 20 to 50 or more independent genetic loci (STR markers). The event that all of these match the father is statistically extremely rare, making the possibility of a coincidental match virtually zero.
• Thanks to the vast accumulation of testing records, the reliability of a probability level of 99.99% is thoroughly backed up in practice.
• Statistical guidelines from the U.S. Department of Justice's National Institute of Justice (NIJ) also state that it is standard practice to conclude "this man is the father" when the probability of paternity is 99.0% or higher [ref:4].

In other words, a figure of 99.99% far exceeds international standards and means that the possibility of another man being the father by coincidence has effectively been ruled out. Across the enormous volume of testing conducted worldwide to date, there have been no reported cases of a false positive at this probability level, and it is treated as "de facto proof" both scientifically and legally [ref:5].

What Is the Difference Between 99% and 99.99%?

At first glance, 99% and 99.99% appear to differ by a mere 0.99%. In reality, however, this "small difference" represents a 100-fold difference in precision.

Let's compare the two more concretely.

Internationally, 99.0% as set by the AABB is treated as the minimum standard for the probability of paternity, but a growing number of reputable testing laboratories in Japan carry out DNA paternity testing with a guaranteed minimum precision of 99.99% — more than 10 times stricter [ref:6]. As a result, DNA paternity testing conducted in Japan can be said to offer reliability that far exceeds international standards.

While a probability of 99% is already quite high, legal proceedings such as court cases generally require an even higher level of precision. A result of 99.99% or higher provides sufficient grounds for a court to recognize that "the existence of a father-child relationship has been scientifically proven."

Why Does It Never Reach 100%?

The reason DNA paternity test results never reach 100% in theory is rooted in the principles of statistics and genetics.

DNA testing does not directly compare a sample against the DNA of every single person on Earth. Instead, testing derives its results using a statistically analyzed "probability of a match." Since the discriminating power of DNA profiling exceeds 2 trillion, the probability that another person within the same population group shares an identical DNA profile is a value extremely close to zero [ref:7].

Even in the extremely unlikely event that another person with an identical DNA profile existed, if the likelihood ratio supporting the alleged father as the biological father is 1,000-fold or greater, it is theoretically considered impossible to overturn the determination that a biological relationship exists with the child.

How the Probability of Paternity Is Calculated

The probability of paternity is calculated through the following steps.

1. Calculating the Likelihood Ratio (LR): For each genetic locus, the ratio is calculated between the probability of observing the DNA type if the specific man is the father, and the probability of observing the same DNA type if an unrelated man is the father.
2. Calculating the Combined Likelihood Ratio: The likelihood ratios obtained at each locus are multiplied together to produce an overall combined likelihood ratio. The more loci analyzed, the larger the final likelihood ratio, and the higher the precision.
3. Setting the prior probability: The prior probability is typically set at 0.5 (50%), and the calculation proceeds from this neutral premise.
4. Deriving the Probability of Paternity (PP): Using the prior probability and the likelihood ratio, the final probability is calculated based on Bayes' theorem.

As is clear from the formula above, because a value of "1" is added to the denominator, it is mathematically impossible for the calculated result to reach 100%. No matter how large the likelihood ratio becomes, the theoretical result will always remain below 100% [ref:7]. However, when the likelihood ratio reaches tens of thousands to millions, the probability of paternity approaches 99.9999% or higher, which in practice provides more than sufficient grounds to conclude that the man is "definitely the father."

Choosing a Testing Laboratory Also Matters!

While the figure of 99.99% carries extremely strong probative value on its own, if a testing laboratory's technical standards or procedural management are inadequate, the reliability of that figure is undermined. To receive trustworthy DNA paternity testing, it is important to choose a laboratory that meets the following conditions.

• International quality certification: Holds ISO 9001 quality management system certification and complies with AABB (American Association of Blood Banks) or CAP (College of American Pathologists) standards, or with the guidelines of the Japanese Society of Legal Medicine [ref:3]
• Strict identity verification and sample management: Verifies identity using photo ID at the time of sample collection, and maintains an unbroken chain of custody throughout the entire process from collection to analysis
• Sufficient number of analyzed markers: Analyzes 20 to 50 or more STR markers, ensuring higher precision
• Result guarantee system: Has a guarantee system for a probability of paternity of 99.99% or higher, with a retesting system in place in the rare case the standard is not met
• Privacy protection system: Holds certification related to personal information protection (such as Japan's Privacy Mark), and manages test results and personal information with strict care

In particular, if you plan to use DNA test results for legal proceedings (such as paternity acknowledgment suits or child custody disputes), it is essential to choose a testing laboratory that can issue a formal test report suitable for submission to a court. It is also important to have pre- and post-test counseling support in place so you can undergo testing with peace of mind.

Cases Where DNA Paternity Testing Is Legally Recognized

DNA paternity testing is used in a variety of legal contexts. A result with a probability of paternity of 99.99% or higher is generally accepted as "de facto proof" in situations such as the following.

• Paternity acknowledgment suits: In lawsuits to legally establish the father-child relationship of a child born to unmarried parents, DNA test results are adopted as conclusive evidence.
• Custody and child support disputes: Used to confirm the existence of a biological parent-child relationship in child custody disputes or child support claims arising from divorce.
• Inheritance matters: Used when it is necessary to prove a blood relationship with a deceased person in connection with an inheritance.
• Correction of birth registrations and family registers: When there is a discrepancy between the family register record and the actual parent-child relationship, correction procedures may be carried out based on DNA test results.

In any of these cases, whether the testing laboratory carries out testing under a quality management system that complies with international standards is a critical factor affecting the evidentiary weight of the results.

Summary

In short, a "99.99% probability of paternity" is a level you can trust with confidence, but where you get tested is the key to that reliability. When you undergo testing at a trustworthy laboratory, that figure functions as unshakeable evidence, both scientifically and legally.

＼Find out who the father of your unborn baby is／

Frequently Asked Questions

Q1. Is a 99.99% probability of paternity really reliable?

A. Yes, it is extremely reliable. A figure of 99.99% means that the chance of another man's DNA coincidentally matching is less than 1 in 10,000. This far exceeds the international minimum standard of 99.0% set by the AABB, and is recognized both scientifically and legally as "de facto proof" [ref:3].

Q2. Why does DNA paternity testing never reach 100%?

A. Because the probability of paternity is derived through statistical calculation based on Bayes' theorem, it is mathematically impossible for it to reach 100%. However, this does not mean the precision is insufficient — a probability of 99.99% or higher provides more than sufficient grounds in practice to conclude that the man is "definitely the father" [ref:7].

Q3. How much difference is there between 99% and 99.99%?

A. The numerical difference appears to be only 0.99%, but the actual precision differs by a factor of 100. At 99%, a coincidental match could occur at a rate of 1 in 100, whereas at 99.99% the possibility drops to 1 in 10,000. For use as legal evidence, a precision of 99.99% or higher is recommended.

Q4. Can DNA paternity test results be used in court?

A. Yes, results from DNA paternity testing conducted at a proper testing laboratory are accepted as evidence in court proceedings such as paternity acknowledgment suits, custody disputes, and inheritance matters. However, submission to a court requires a formal test report that includes strict identity verification and chain-of-custody records for the samples. We recommend using a laboratory that complies with ISO certification and AABB standards.

Q5. What points should I check to choose a reliable testing laboratory?

A. When choosing a reliable testing laboratory, check the following points: (1) holds international quality certification such as ISO 9001; (2) complies with international standards such as AABB or CAP; (3) analyzes a sufficient number of STR markers, 20 or more; (4) has a guarantee system for a probability of paternity of 99.99% or higher; and (5) has privacy protection measures in place (such as Japan's Privacy Mark) [ref:3].

Q6. Can DNA paternity testing be done before birth?

A. Yes, DNA paternity testing can be performed even before birth. There is a non-invasive method (NIPPT) that uses fetal-derived cell-free DNA (cfDNA) contained in the mother's blood, allowing testing to be carried out during pregnancy without placing any burden on the mother or fetus [ref:1]. Please feel free to contact seeDNA for more details.

[References]