The ABO system is the classic teaching example of codominance, and it is one of the few human traits where a simple single-locus model genuinely works most of the time. Three alleles, four phenotypes, clean arithmetic. The blood type calculator above runs that model in both directions and adds the Rh factor as a second, independent gene.
Arb Digital publishes free science calculators for coursework and general curiosity. This one comes with a firm boundary attached, stated here rather than buried at the bottom: it is an educational model of inheritance. It is not a paternity test, it is not diagnostic, and it must not be used to make any medical, legal or family decision. Real blood typing is done on a sample in a laboratory, and there is no substitute for that.
What This Blood Type Calculator Does
In forward mode it takes two parent types and returns the probability of each of the eight combined ABO and Rh outcomes in a child. In reverse mode it takes one parent and a child and returns every type the second parent could have, which is the more interesting calculation because it is the one that actually rules things out.
Where you know a parent's genotype rather than just their group — because a previous generation or an earlier child settled it — you can select it, and the answer sharpens. Where you do not, the tool weights the two possible genotypes equally and says so in the results line. That is an assumption of convenience, not a population measurement, and real allele frequencies vary a great deal between populations.
ABO and Rh are computed independently because they sit on different chromosomes: the ABO gene on chromosome 9 and the RHD gene on chromosome 1. Independent assortment means the combined probability of, say, A positive is simply the ABO probability multiplied by the Rh probability. If you want to see any single-gene cross laid out as a grid instead, the Punnett square calculator draws the squares.
How to Use It
- Choose the direction first. Forward mode wants two parents. Reverse mode reads the second box as a child rather than as a parent.
- Pick the ABO group. Use the plain A or B option unless you genuinely know the genotype; the AA and AO entries are for when a previous cross has already revealed it.
- Set the Rh factor. Negative is unambiguous because it needs two copies of the non-functional allele. Positive could be one copy or two.
- Read the full list of outcomes, not just the headline. In forward mode every non-zero probability is a real possibility.
- In reverse mode, look at what is excluded. The exclusions are the informative part; the possibilities usually are not.
How ABO and Rh Inheritance Are Calculated
The ABO locus has three common alleles. As the ABO Blood Group chapter in the NCBI MINI Medical Genetics Summaries puts it, the ABO blood type is inherited in an autosomal codominant fashion: the A and B alleles are codominant with each other, and the O allele is recessive to both. So AA and AO both give group A, BB and BO both give group B, AB gives group AB, and only OO gives group O.
Rh is simpler in the model. The D antigen is dominant, so DD and Dd are both Rh positive and only dd is Rh negative. Two Rh positive parents can have an Rh negative child if both carry a d allele; two Rh negative parents cannot have an Rh positive child in this model.
A worked example matching the page defaults. An A parent of unknown genotype crossed with a B parent of unknown genotype, both Rh positive with unknown genotype. Each A parent is modelled as half AA and half AO, so it passes an A allele 75 percent of the time and an O allele 25 percent. The same holds for B. The child is therefore AB with probability 0.75 × 0.75 = 56.25 percent, group A at 18.75 percent, group B at 18.75 percent, and group O at 6.25 percent. Every one of the four groups is possible from this pairing, which is what makes A crossed with B the least informative combination in the whole system. On the Rh side, each parent passes D 75 percent of the time, so the child is Rh negative 6.25 percent of the time. The single most likely outcome is AB positive at 52.7 percent.
Why This Is Not a Paternity Test
Blood groups can occasionally exclude, and they can never confirm. That asymmetry is the entire point and it is routinely misread in both directions.
Exclusion happens when a child has a group that the claimed parents cannot jointly produce — a group AB child from a group O parent, for example. Even then, the exclusion is only as good as the model, and the model has documented exceptions described below. Confirmation never happens at all. Group A is carried by a large fraction of the population, so an A child and an A man tells you nothing whatsoever; the same result would hold for millions of unrelated men.
Modern parentage testing uses short tandem repeat markers, examining a panel of highly variable regions and producing probabilities that run to many nines. ABO carries a trivial amount of information by comparison. Historic legal cases that turned on blood groups did so because nothing better existed, not because the method was sound. Nothing on this page should be used to reason about parentage, and nobody should draw a conclusion about a real family from a probability produced by a coursework model.
The Documented Exceptions the Simple Model Misses
Three exceptions are worth knowing, because they are the reason a laboratory result occasionally contradicts the family tree.
The first is the Bombay phenotype. The A and B antigens are built by modifying a precursor called the H antigen, which is made by a separate gene. The NCBI reference chapter on the ABO blood group in Blood Groups and Red Cell Antigens explains that individuals homozygous for null alleles at that H locus do not produce H antigen at all, and describes this rare phenotype as the Bombay phenotype. Such a person types as group O on a routine test no matter what ABO alleles they carry, so they can pass an A or B allele to a child while appearing to be O themselves.
The second is cis-AB, a rare variant in which a single chromosome carries a transferase that produces both A and B antigens. Because both are inherited together as one unit, a cis-AB parent can have a group O child, which the ordinary model forbids.
The third is weak D, on the Rh side. Some people express so little D antigen that they type as Rh negative by one method and Rh positive by another, and the classification can differ between a donor centre and a hospital laboratory because the two use different thresholds for different reasons. That alone can make a family's Rh results look impossible when nothing unusual is happening at all.
What Reverse Mode Is Actually Good For
Working backwards from a child sounds more powerful than it is, and seeing why is a good lesson in how evidence works. Take a group O child with a group O parent. The second parent must have passed an O allele, so they can be A, B or O, and only AB is ruled out. One possibility eliminated out of four.
Now take a group AB child with a group A parent. The second parent must have passed a B allele, so they are B or AB. Two possibilities eliminated, which is as informative as this system gets. The pattern is general: rare child groups exclude more than common ones, and a group AB or group O child tells you more than a group A child does. The Bayes theorem calculator is the right tool if you want to think properly about how much a piece of evidence like this actually shifts a belief, and the probability calculator handles compound questions across several children.
Why Rh Negativity Matters Clinically
The Rh factor is not just a letter after the group. When an Rh negative person is exposed to Rh positive red cells, the immune system can produce antibodies against the D antigen. That matters in transfusion, and it matters in pregnancy where an Rh negative parent may carry an Rh positive fetus.
This is precisely where a calculator stops and medicine starts. Management of Rh incompatibility is a clinical matter with an established standard of care, decided by testing and by a clinician who knows the case. A probability from a teaching model has no role in it whatsoever. If Rh status is relevant to a pregnancy or a transfusion, the answer comes from a laboratory and from a doctor, not from this page or any other.
Arb Digital publishes hundreds of free calculators across biology, maths, physics and finance — no sign-up, no limits. If something you need is missing, tell us and we will look at building it.
Browse All Free Tools Suggest a ToolCommon Mistakes to Avoid
- Treating a match as confirmation — blood groups can sometimes exclude a relationship and can never establish one, because common groups are shared by an enormous number of unrelated people.
- Forgetting that A and B hide an O allele — a group A person may be AA or AO, and the difference decides whether a group O child is possible.
- Assuming the model has no exceptions — the Bombay phenotype, cis-AB and weak D all produce family patterns that the simple model calls impossible.
- Combining ABO and Rh incorrectly — they are separate genes on separate chromosomes, so the combined probability is the product of the two, not a shared calculation.
- Using a modelled probability where a test belongs — transfusion, pregnancy management and parentage all require laboratory results, never an inheritance calculation.
Related Free Tools From Arb Digital
See any single-gene cross drawn as a grid with the Punnett square calculator, look at a polygenic trait handled honestly with the baby eye color calculator, and work through compound chances using the probability calculator. The Bayes theorem calculator shows how much a single piece of evidence really moves a belief, the expected value calculator averages across outcomes, and the DNA to mRNA converter covers transcription for coursework. The full free online tools hub lists everything else.
Frequently Asked Questions
Not in the standard model, because group O parents carry only O alleles and can pass nothing else. The documented exceptions are the Bombay phenotype, where a person types as O but carries an A or B allele, and the rare cis-AB variant.
All four. If both carry a hidden O allele, the child can be AB, A, B or O. With equal weighting of the unknown genotypes that comes to roughly 56 percent AB, 19 percent A, 19 percent B and 6 percent O.
No, and it must never be used to try. Blood groups can occasionally exclude a claimed relationship and can never confirm one. Parentage testing uses panels of highly variable DNA markers and is a laboratory procedure.
Yes, if both carry the negative allele. Each would pass it a quarter of the time, giving the child a one in four chance of inheriting two copies and typing as Rh negative.
It is a rare condition in which a person cannot produce the H antigen that the A and B antigens are built from. They type as group O on a routine test regardless of which ABO alleles they carry, so they can pass an A or B allele to a child.
Because group A and group B each cover two genotypes. Knowing which one a parent has sharpens the answer considerably. Where it is unknown, the tool weights the two equally, which is an assumption rather than a measurement.
No. The ABO gene sits on chromosome 9 and the RHD gene on chromosome 1, so they assort independently. The chance of a combined type is the ABO probability multiplied by the Rh probability.
Usually because of one of the documented exceptions. The Bombay phenotype, cis-AB and weak D all produce results the simple model calls impossible, and weak D in particular can be classified differently by different laboratories.
This calculator is an educational tool that models the common ABO and Rh alleles for coursework and general interest. It is not a paternity test, it is not diagnostic, and it must never be used for any medical, legal or family decision. Documented exceptions including the Bombay phenotype, cis-AB and weak D fall outside the model entirely. A person's real blood group is established by laboratory testing of a sample, and any question about transfusion or pregnancy belongs with a qualified clinician.
