3 ms·
> 78% identical seems like a lot more than the expected 50%. How likely is it for two sperm to be 78% identical? Statistics doesn't work that way. Each chromos
by gwbas1c 8y ago
> 78% identical seems like a lot more than the expected 50%. How likely is it for two sperm to be 78% identical?
Statistics doesn't work that way. Each chromosome has a 50% chance of being shared between both children. It's possible, but unlikely, that each child would have all the same chromosomes from the father; or share none of the chromosomes from the father.
Random events have no memory. Meaning, if you flip a coin and get heads, the coin has no memory that makes it more likely to get tails on the second flip.
Thus, while each chromosome has a 50% chance of being shared, there is no "memory" that ensures that both children share half of their father's chromosomes.
- thaumasiotes 8y ago> Statistics doesn't work that way. I asked how likely it was for two sets of gamete chromosomes to be similar at levels a particular distance from the expected mean similarity. That's pretty much exactly the kind of question statistics handles. > Each chromosome has a 50% chance of being shared between both children. On the other hand, biology really doesn't work this way. Chromosomes recombine in meiosis; you shouldn't see a child sharing any entire chromosome, except the father's sex chromosome, with either parent.
- SamBam 8y agoTypically there are only 3-4 crossovers per chromosome. This means that large chunks of DNA get copied together, and as such the assortment of every base is not as simple as assuming each has an independent chance of ending up in a gamete. Rather, nearby bases are extremely likely to go into the same gamete. This is why it's quite possible to get gametes that are similar in quantities much higher than the "expected" 50%. Instead of being 3 billion items randomly assorted, it's roughly 22 chromosomes x 4-5 chunks = 100 of so chunks of DNA, albeit split at random locations. This simply spreads out the "normal" distribution. 50% is still the expected similarity of any two gametes, but it's not surprising for any two to be much more similar, or much less similar.