As a top paternity lab we often get asked, “If I’m considered the biological father, why do my test results show 99.99% and not 100% probability of paternity?” That’s an excellent question, and it’s also a scientific one. Here’s an explanation in layman’s terms about why probability of paternity can never—that’s correct—never be 100%.
How a Baby Inherits DNA from its Parents
During the reproduction process, the DNA of two individuals together—mother and father—randomly produces a very unique combination of genetic markers in each cell. This combination of markers is unique to the new person they create together.
The genetic material of a child is inherited from the parents in equal portions, hence the child’s genome as presented in a paternity report reflects the 50% DNA they get from their mother and the 50% DNA they get from their father.
In order for the man tested to be considered the biological father of the child tested, there must be a match between them at each one of the locations tested, although there may be an exception made for a known genetic mutation. So if a man matches the child at each location tested, shouldn’t the probability of paternity be 100%? No, and this is where the science comes in.
Why Probability of Paternity Can Never be 100%
Comparing the DNA sequence of a child to a parent can identify if one of them was derived from the other. This does not mean that we can achieve a probability of 100%.
When considering the possibility of paternity, specific sequences are reviewed and ethnic data are used to calculate a probability of paternity. Analysis required for paternity testing is all about statistics and the only way a man could ever be considered the biological father with 100% probability is if every man in the world with the same ethnic background were also tested. Of course, this is impossible!
Basics of Calculating Probability of Paternity
Paternity Index: The probability of match between alleged father and child at each genetic location has its own calculation, called a Paternity Index (PI). The strength of the match at that location is determined by a variety of factors.
Combined Paternity Index: We generally test a minimum of 20 genetic locations, and—as mentioned—each has its own associated paternity-index value. The PI are multiplied together to get what’s called a Combined Paternity Index (CPI). And that number is the statistical probability of relationship. In the example below, the CPI is 1,716,973,979. This could read: The man tested is 1,716,973,979 times more likely to be the biological father than an unrelated, untested man with the same ethnic background.
Using the strength of the genetic results in the Combined Paternity Index, a Probability of Paternity is calculated. We’ve already established that number can never be 100% unless every man in the world can also be tested. But the higher the CPI, the higher the Probability of Paternity. Anything over 99% is considered conclusive. This example shows an extremely high CPI; the minimum for a CPI to be considered conclusive can be as low as 100 for an at-home test and 200 for a legal test.
But What if there is a Mismatch at a Genetic Location?
If there is no match at one (or more) of the genetic locations tested and the mismatch isn’t due to a known mutation, the PI for that location is 0 %. This means that when all PIs are multiplied together, that one 0% makes the Combined Paternity Index 0%, and the man is not considered the biological father.
If the DNA of two individuals don’t match, then we can say with 100% certainty that a man is not the biological father of a child.
So What Does This All Mean?
A 99.99% probability of paternity is so powerful, that it’s routinely accepted as evidence in favour of paternity by genetic scientists, courtroom judges, and the Ministry of Justice. The upshot is: A DNA paternity test probability of 99.99% is strong enough for a judge to confidently allow (or deny) child support, immigration, or even conviction in a criminal case.