Every paternity test report you get back from a lab has a table on it. Down the left side is a column of codes that look like typos: D3S1358, TH01, D18S51, FGA, vWA. Next to each code are two small numbers for the alleged father, two for the child, and often two for the mother. At the bottom of the page is a single number that most people skip straight to: the probability of paternity.
Those codes are the genetic markers the lab checked. They are the whole reason the final number at the bottom of the report is trustworthy. USDC's home paternity test kit can analyze up to 28 unique autosomal markers in total, but the lab does not run all 28 on every case. This article walks through what those markers actually are, what the standard workflow looks like for a paternity test, when the lab reaches for a second panel, and how to read the table when your envelope arrives.
Fact 1: STR stands for Short Tandem Repeat, and it is just a pattern that repeats in your DNA
A short tandem repeat, or STR, is a stretch of DNA where a short pattern (usually 2 to 6 letters) repeats back-to-back several times in a row. Think of it as the same short word written over and over: CATG-CATG-CATG-CATG. At one location in your genome you might have 12 of those repeats. At the same location in someone else's genome, there might be 8 repeats. In someone else, 17.
The number of repeats at each STR location is inherited. You got one count from your biological mother and one count from your biological father. Because the number of repeats varies widely from person to person, STRs are the workhorses of paternity, kinship, and forensic DNA testing. The National Institute of Standards and Technology's STR database catalogs the loci human identity labs use and the range of repeat counts observed across populations.
When your report lists D3S1358 with numbers "15, 16" next to the child, that means the child inherited a version of that STR with 15 repeats from one parent and a version with 16 repeats from the other.
Fact 2: USDC's lab can analyze up to 28 genetic markers, and the standard panel runs 23 of them for statistics
Every paternity test USDC runs starts with the same standard panel. On your report you will see roughly 25 marker labels, but not all of them do the same job.
- 23 autosomal STR markers — these are the ones that carry the statistics. They live on your regular non-sex chromosomes and are inherited from both biological parents. Every one of them contributes a factor to the Combined Paternity Index (CPI) and the Probability of Paternity at the bottom of your report.
- A sex-typing marker — this marker tells the lab whether each sample came from a male or a female donor. It is reported on the sheet but does not contribute to the CPI or Probability of Paternity because it is not informative for parentage.
- An internal quality-control marker — the lab uses this one to verify sample integrity. It is not reported to customers and is not included in any statistic.
So when someone says "the lab tested 25 markers," what they mean is 23 autosomal STRs that do the statistical work, one sex-typing marker, and one internal QC marker. The 23 autosomal markers are what actually decide the answer.
Fact 3: For a standard paternity test, the 23-marker standard panel is far more than enough
Consumer paternity testing hit "definitive" a long time ago. Twenty markers is enough to reach a mathematically definitive result in the vast majority of cases. USDC's lab can analyze up to 28 genetic markers and runs 23 autosomal STR markers as the standard panel on every paternity test, which pushes past the industry baseline comfortably. In a routine case with a clean sample and a straightforward biological relationship, the standard panel produces a Combined Paternity Index in the tens of millions or higher and a Probability of Paternity of 99.99% or greater.
There is a common assumption that a higher marker count always translates to a more reliable answer. In practice, once you are past 20 well-chosen autosomal STRs, the probability number stops meaningfully improving. The difference between 99.99% and 99.999% is not something that changes what the report is telling you. The result is the same either way. So for the ordinary paternity case — one alleged father, one child, one mother if she is available — the standard 23-marker panel by itself gives you an answer you can rely on.
The one situation where marker count still matters for paternity testing is mutations. A mutation is a change in a specific STR that is passed from parent to child but does not match the parent's own repeat count exactly. Mutations are rare per locus but can happen. If a paternity test shows a mismatch at just one or two markers while every other marker matches perfectly, the report has to answer the question: is this an inherited mismatch (which would suggest exclusion) or is it a mutation (which would still be consistent with paternity)? More markers make that distinction easier. The standard 23 autosomal markers give the lab enough statistical power to sort this out in the overwhelming majority of cases without needing to run additional tests.
Fact 4: Kinship testing reports a different statistic (CRI), and the math is harder
Kinship tests — the tests you order when the alleged father is not available and you have to work with a paternal grandparent, an aunt or uncle, or a sibling — do not report a Probability of Paternity number. They report a Combined Relationship Index, or CRI. USDC uses the standard three-tier CRI framework:
- CRI greater than 10 — the tested relationship is supported by the DNA
- CRI between 0.1 and 10 — the result is inconclusive
- CRI below 0.1 — the tested relationship is not supported
Kinship math is harder than paternity math for a simple reason. In a paternity test, the alleged father contributes half the child's DNA directly. In a grandparent, aunt/uncle, or sibling test, the shared genetic material is one step further away. A grandparent shares about 25% of DNA with a grandchild on average. Half-siblings share about 25%. Aunts and uncles share about 25%. Those percentages are averages — actual shared segments vary case to case — and they leave less signal for the standard panel to work with.
USDC's lab can analyze up to 28 genetic markers, and in the majority of kinship cases the 23 autosomal markers in the standard panel produce a decisive CRI on the first pass. But not every kinship case is decisive at 23 markers. Sometimes the CRI lands in the inconclusive middle range, and the lab needs more discrimination to push the number into a supported or not-supported band.
Fact 5: When kinship is inconclusive, the lab adds a second panel to reach up to 28 unique autosomal markers
This is where the "up to 28 markers" figure comes from. If a kinship test's initial run on the standard panel produces a CRI in the 0.1 to 10 inconclusive range, the lab does not report an inconclusive result and stop there. It adds a second amplification panel that analyzes an additional five autosomal STR markers the standard panel does not cover. Those five loci do not overlap with the standard panel, so the result is 23 plus 5 equals 28 unique autosomal markers.
Those extra five markers are chosen because they add independent statistical information. Adding them to the CRI calculation almost always moves an inconclusive result decisively into either the supported (CRI greater than 10) or not-supported (CRI less than 0.1) band. In the small remaining percentage of cases where even 28 markers cannot resolve the relationship, the lab may recommend testing an additional relative to add a fresh line of comparison.
So the practical meaning of "up to 28 markers" is straightforward: the 23 autosomal markers in the standard panel do the work on paternity tests and on most kinship tests, and up to 5 additional markers are added for kinship tests that did not resolve on the first run. Paternity tests do not get the extra 5 markers because they do not need them.
Fact 6: How markers show up on your report is simpler than it looks
When your USDC report arrives, the marker table follows a predictable layout. It looks roughly like this, column by column:
- Locus: the code for the marker (D3S1358, TH01, D18S51, etc.).
- Alleged Father (or tested relative for a kinship test): two numbers, one for each version of that marker he carries.
- Child: two numbers, one inherited from the biological mother, one from the biological father.
- Mother (if her sample was included): two numbers.
- Paternity Index (or Relationship Index on a kinship report): the statistical weight this specific locus contributes.
The two numbers per person at each locus are called alleles. You have two copies of every autosomal chromosome (one from each biological parent), so you have two alleles at every autosomal STR marker. The child's two alleles at each locus should match one allele from the biological mother and one from the biological father. When the alleged father's alleles include the one the child inherited from him, that locus contributes a positive index. When they do not, that locus contributes an exclusion. The MedlinePlus genetics glossary is a good plain-language reference if any of this vocabulary is new.
The sex-typing row on your report will show letters instead of numbers — an X paired with either another X (female) or a Y (male). This is just the sex-verification check the lab ran to confirm the samples came from who you said they came from.
The final probability number at the bottom of a paternity report (Probability of Paternity), or the CRI on a kinship report, is not a separate test result. It is a mathematical summary of every autosomal row above it.
Frequently Asked Questions
Do more markers always mean a more accurate paternity test?
Not really. USDC's lab can analyze up to 28 genetic markers, and once a paternity test uses 20 or more well-chosen autosomal STRs, the answer stops changing meaningfully. The 23-marker standard panel gives you Probability of Paternity numbers of 99.99% or higher in ordinary cases. Running more markers would tighten the statistical confidence a small amount further but would not change the yes-or-no answer.
Why do you run 28 markers on kinship tests but not paternity tests?
Paternity math is easier because the alleged father is one biological step from the child. Kinship math has to work across a wider genetic gap (grandparent to grandchild, uncle to niece, sibling to sibling), so the standard 23-marker panel sometimes leaves the result in an inconclusive range. When that happens, adding up to 5 additional markers from a second panel almost always pushes the result into a decisive range. Paternity tests do not run into that problem, so the extra 5 markers are not needed.
What is the sex-typing marker on my report, and why is it there if it does not affect the answer?
The sex-typing marker is the one labs use to verify the biological sex of each sample. It appears on your report because it confirms the samples were collected from the right people. It is not included in the Combined Paternity Index or in the Probability of Paternity because it does not carry information about the biological relationship — it just tells you the sample came from a male or a female donor.
Do we need the mother's sample for the standard 23-marker panel to work?
No. A motherless case (just alleged father and child) still runs the full standard panel and can still return a definitive Probability of Paternity. Including the mother's sample sharpens the interpretation at some loci by removing ambiguity about which of the child's alleles came from which parent, but it is not required. If a kinship test needs the additional 5 markers, the mother's sample can be added at that stage if it was not part of the original submission.
Is a 23-marker or 28-marker home test court-admissible?
The marker count is a separate question from court admissibility. Home paternity tests, regardless of how many markers the lab runs, are not court-admissible because the sample collection is not witnessed and there is no chain of custody. If you need a court-admissible result today, ask an accredited testing provider in your area about legal paternity testing with a witnessed collection.
0 comments