A low number statistics probability-problem (genetics)

  • Context: Undergrad 
  • Thread starter Thread starter B0b-A
  • Start date Start date
  • Tags Tags
    Genetics Statistics
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
Sign up Log in
5 replies · 2K views
B0b-A
Messages
155
Reaction score
32
A probability problem (genetics), not homework, honest.

3 of 5 siblings are affected by an unknown inherited disease*.

That this outcome is higher than 1 in 4 which is typically the case in recessive traits suggests that the disease is more likely to be dominant than recessive.

Given 3 of 5 siblings are affected how much more likely is it that this genetic illness is dominant rather than recessive ?.

Like I said this isn't homework so I can't confirm if any answer given is correct.

[ *adult onset : inheriting the trait would not make their birth less likely ].
 
Physics news on Phys.org
Without knowing the status of the parents, I don't think this problem works. In particular, the 1/4 is expected only in a case where a dominant trait would not appear at all.

With assumptions about their health, you can calculate the likelihood for both hypotheses and compare them.
 
mfb said:
Without knowing the status of the parents, I don't think this problem works.

In this case only one of the parents of the five siblings is affected (symptomatic).
(sorry forgot to mention that :redface: ).
 
Then there is no reason to expect a rate of 1/4.

Based on the information that neither zero nor all childen got symptoms, you can determine the genetic composition of both parents in both cases and solve the problem.
 
mfb said:
Then there is no reason to expect a rate of 1/4 ...

OK now I see 1in 4 is only expected in recessive inheritance where both parents were unaffected carriers: both carry one copy of the disease gene. But in this case one parent was affected.

mfb said:
... Based on the information that neither zero nor all childen got symptoms, you can determine the genetic composition of both parents ...

If it were dominant the affected parent cannot have two copies of the disease gene, otherwise all the children would be affected.

If recessive the unaffected parent must have a copy of the disease gene, otherwise no children would be affected.

mfb said:
... you can determine the genetic composition of both parents in both cases and solve the problem.

Now my possible punnet squares are giving me 1 in 2 odds that a child would be affected, whether it was dominant or recessive inheritance.

So that it affected 3 of 5 siblings doesn't make it more likely to be dominant than recessive (?)
 
Last edited:
Correct.

You could argue that the information "some do not get the trait, some do" itself is some information, but then you need some estimate about the total rate of the trait in the population.
 
  • Like
Likes   Reactions: 1 person