My grandma is O- blood type, my grandpa is B+, but my mother is A+ and my aunt is AB+.

Is that possible?

I've just looked at different sources like this and all of them are saying that from O and B, comes only B or 0, and A and AB are forbidden.

How can my mother be a child of my grandparents and how my aunt can be a child of my grandma at all? Or can we 100% say that they are not the real children of those parents, assuming that all blood types are correct?

  • $\begingroup$ maybe your grandmother is not really O negative or your grandfather is not B positive $\endgroup$ – Vance L Albaugh Jul 21 '16 at 11:41
  • $\begingroup$ Maybe, but I doubt so. Because my grandma was hospitalized (including surgeries) multiple times, and my grandpa was tested for his driving license. $\endgroup$ – judoka_acl Jul 21 '16 at 12:15

It is correct that O- and B+ parents cannot produce biological children with A+ and AB+ blood groups, with the exception of a few rare edge cases.

Firstly, as noted in the other answer, your grandmother may have the Bombay phenotype. However, this is unlikely, due to the fact that she has been previously hospitalised and it was not noted.

Another unlikely possibility is that your grandmother is a genetic chimera. This means she would have two sets of genes, and her bone marrow hematopoietic cells had the O- genotype and phenotype, while her reproductive cells had an A allele.

If we exclude the rare edge cases, then it is strong evidence that your mother and aunt are not the biological children of your grandparents, and you should consider performing genetic testing to verify this if you wish.

  • 2
    $\begingroup$ Wouldn't a wikipedia link to genetic chimera be better than a link to a clickbait site? $\endgroup$ – Pyritie Jul 21 '16 at 15:30
  • $\begingroup$ @Pyritie I thought the "clickbait" site had more detail and was better as a case study, but the wikipedia page works too. $\endgroup$ – March Ho Jul 22 '16 at 0:06

Assuming true paternity and correct ABO blood type detection, your grandma could likely have a Bombay phenotype (h/h blood group), which would make her blood look like the O type: http://www.ncbi.nlm.nih.gov/books/NBK2268/

This means that she would be double recessive to (and therefore lack) the H antigen, which is a precursor to A and B antigens. She would actually be AO or AA (genetically speaking) when it comes to the ABO group.

EDIT: As noted in the comments, this is indeed an extremely rare blood type, troublesome for transfusions, and should have been noticed prior to surgery. Other hypotheses are of course entirely plausible (another father, chimerism, etc.). But I think we can't discard it yet until the OP does some more extensive testing, as the transfusion issue could be handled by autologous blood transfusion or a very well stocked blood bank. Alternatively, the phenotype could be Para-Bombay, meaning that a little H antigen is produced and turned into A/B; this could be slightly more manageable for transfusions.


  • 1
    $\begingroup$ Maybe unlikely since, the OP mentions surgeries and Bombay blood is difficult to procure. $\endgroup$ – Polisetty Jul 21 '16 at 12:27
  • $\begingroup$ Isn't that actually like to say that my grandma group wasn't correctly determined? Or you meant that in the case of Bombay phenotype it's not possible at all to be correctly determined according to the ABO group? $\endgroup$ – judoka_acl Jul 21 '16 at 12:28
  • $\begingroup$ @lalamer Someone with Bombay phenotype will appear as O in standard ABO testing regardless of their actual genotype. This is known as epistasis $\endgroup$ – March Ho Jul 21 '16 at 12:32
  • $\begingroup$ Is there any other way except this Bombay phenotype? $\endgroup$ – judoka_acl Jul 21 '16 at 12:37
  • 2
    $\begingroup$ @lalamer (and expanding on the comment by March Ho) ABO typing is firstly done with specific antibodies for A and B antigens (anti-A and anti-B), which induce clumping of the blood accordingly; i.e., A blood is clumped by anti-A, B by anti-B, AB by both, and O by none. If there is no H, then neither A nor B are produced or detected by the basic test; you would have to verify this by adding anti-H. An additional note: ABO and H are synthesized by two different enzymes encoded by separate genes (ncbi.nlm.nih.gov/gene/28 and ncbi.nlm.nih.gov/gene/2523). $\endgroup$ – narvalkiria Jul 21 '16 at 13:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.