Learning how to read paternity test results comes down to three things on the page: the allele table that compares DNA at each tested location, the statistics that summarize those comparisons (Paternity Index, Combined Paternity Index, and Probability of Paternity), and the plain-language conclusion at the bottom stating whether the alleged father is included or excluded. Everything else on the report supports one of those three.
First, Check What Kind of Report You Have
Before reading the numbers, confirm whether your report is from a legal test or an informational (at-home) test. Both use the same science and produce the same kinds of numbers. The difference is chain of custody. A legal test is collected by a trained third party at a certified facility, with identity verification and sealed samples, and can be submitted to a court for child support, custody, immigration, or a birth certificate change. An informational test uses swabs collected at home without identity verification, so most courts will not accept it even though the genetics are equally accurate.
For any legal use, the lab generally needs accreditation from the AABB (formerly the American Association of Blood Banks), which enforces validated quality standards and documented chain of custody.1AABB. DNA Relationship Testing FAQs The U.S. State Department specifically requires an AABB-accredited lab for DNA testing in citizenship or immigration cases.2U.S. Department of State. Information for Parents on U.S. Citizenship and DNA Testing If you plan to use the report for anything official, confirm the lab’s accreditation before you rely on the numbers inside.
Check the Identifying Section
The top of the report lists names, dates of birth, and the sample collection date for each person tested: the alleged father, the child, and the mother if she participated. Samples are almost always buccal (cheek) swabs. On a legal test, this section also records who collected the samples and where. Read it carefully. A wrong name or date here is enough to create problems if the report ever goes to a court or agency, and it is far easier to fix at the lab than after the fact.
How to Read the Genetic Marker Table
The main body of the report is a table of genetic markers, also called loci. Each locus is a specific spot on your DNA where the code varies between people. AABB standards require at least 20 loci, and many labs test more.3AABB. Proposed Standards for Relationship Testing Laboratories Columns typically show the locus name, the child’s two alleles, the mother’s alleles (if she was tested), and the alleged father’s alleles. Alleles are shown as numbers.
Every person carries two alleles at each locus, one inherited from the mother and one from the biological father. To read a row, look at the child’s two alleles. One should match an allele the mother carries. The other one, called the obligate paternal allele, must match one of the alleged father’s two alleles at that locus. If the alleged father’s profile lines up with the obligate paternal allele at every locus, the genetic evidence strongly supports paternity.
A mismatch at a single locus does not automatically mean exclusion. Genetic mutations happen naturally during reproduction, and a single-locus mismatch is more often a mutation than a true exclusion. Labs generally require mismatches at two or more independent loci before calling an exclusion.4National Institute of Justice. Mutation and Paternity When a single mismatch shows up, the lab will usually flag it and may run additional markers.
What the Three Statistics Mean
Under the allele table you will see three numbers that translate the raw comparisons into something usable: the Paternity Index at each locus, a Combined Paternity Index for the test as a whole, and a Probability of Paternity expressed as a percentage.
Paternity Index
Each locus where the alleged father matches the obligate paternal allele gets its own Paternity Index (PI). The PI at that locus answers one question: how many times more likely is this genetic match if the tested man is the father than if a random unrelated man from the same population is the father? A PI of 5 at a locus means the match is five times more likely under paternity than under a random-man scenario. Your report will list a PI value for every locus in the table.
Combined Paternity Index
The Combined Paternity Index (CPI) multiplies all the individual PI values together across every locus. Because it compounds evidence from 20 or more markers, the CPI can reach very large numbers. A CPI of 1,000,000 means the genetic evidence is one million times more likely if the tested man is the biological father than if a random man from the same population is.
CPI calculations rely on population frequency databases that track how common each allele is within specific ethnic or racial groups, because allele frequencies differ across populations.5PubMed Central. Choice of Population Database for Forensic DNA Profile Analysis Your report may name the database used or show calculations against more than one. The CPI can shift slightly depending on the database, but in clear-cut cases the differences are negligible.
Probability of Paternity
The Probability of Paternity (POP) converts the CPI into a percentage. Labs start from a “prior probability” of 0.5, meaning that before any DNA is looked at, the tested man is assumed to have a 50/50 chance of being the father.6National Institute of Justice. Probability of Paternity That neutral starting point makes sure the DNA drives the result. The formula: POP equals CPI divided by (CPI + 1), times 100.
A CPI of 100 works out to about 99.01%. A CPI of 1,000,000 works out to 99.9999%. You will never see exactly 100%, because the statistical model always leaves a theoretical possibility that another man could match. In practice, any POP above 99% is strong genetic evidence of paternity, and most inclusion reports show 99.99% or higher.
The Bottom-Line Conclusion
At the end of the report, the lab states a plain-language conclusion. There are three possible outcomes.
Inclusion (Not Excluded)
An inclusion is usually worded as “the alleged father is not excluded as the biological father.” The double negative sounds hedging, but it is the scientifically correct way to say the DNA is consistent with paternity. AABB standards require a CPI of at least 100 to 1 before a lab can report genetic evidence supporting paternity, which corresponds to a POP of roughly 99% or greater.3AABB. Proposed Standards for Relationship Testing Laboratories Most reports far exceed that minimum.
Exclusion
An exclusion states “the alleged father is excluded as the biological father.” This means mismatches appeared at multiple loci, and the DNA is not consistent with paternity. Exclusions are definitive in the vast majority of cases. Worth knowing, though: the traditional practice of treating the probability as automatically 0% whenever two or more loci mismatch has been criticized by geneticists, whose research has shown that even with two or three inconsistent loci, the true probability of paternity can remain meaningful when mutations are plausible.7International Society for Forensic Genetics. Multiple Mutations, Covert Mutations and False Exclusions in Paternity Testing If your exclusion rests on only two or three mismatches and the allele differences at those loci are small (a shift of one repeat unit), it is reasonable to ask the lab about mutation analysis.
Inconclusive
Inconclusive means the lab could not decide either way with confidence. Under AABB standards, a likelihood ratio between 0.1 and 10 falls in the inconclusive range.3AABB. Proposed Standards for Relationship Testing Laboratories Common causes are a degraded or insufficient sample, an unresolvable mutation pattern, or two closely related men who are both possible fathers. Labs typically recommend retesting with fresh samples or bringing additional parties into the test.
Situations That Can Skew the Reading
Standard testing compares the alleged father’s DNA against a random unrelated man from the same population. That assumption breaks down when close relatives are involved. Brothers usually have profiles distinct enough that testing works, but the lab should know about the relationship so it can run additional markers. Father-son pairs are riskier, because a son inherits 50% of his DNA from his father, and if only one of the two men is tested, a false inclusion is possible. Identical twins are the extreme case: their DNA is effectively the same, and standard testing cannot separate them. If any of these apply to your family, the lab needs to know before the test runs.
Many reports are generated without the mother’s sample, and the results are still reliable. Including her sample makes the statistics stronger, because the lab can identify the obligate paternal allele at every locus with certainty rather than inferring it. The practical effect is a higher CPI and a higher POP, sometimes pushing a Probability of Paternity from 99.99% to 99.99999%. If your report was run without the mother and you need the strongest possible result for court, adding her sample is the simplest way to boost the certainty on the page.