Hacker News Reader

Comment by ronaldx | Hacker News Reader

This article doesn't totally add up.

If the mother was chimeric, she would be chimeric with a 'twin' from the same (grand)mother and father. Therefore her 'twin' chimera would share 50% of her DNA, as any regular sibling.

This would be identifiable on a DNA test as a close relative: she would be genetically equivalent to an aunt rather than a mother.

It's also surprising that they had to test the thyroid when the gametic cell line is the important one (wikipedia suggests they tested her cervix).

View on HN

If the tests were reported to that level – and we're talking 2002 here – the fact that the child appeared to be from another relative wouldn't necessarily put the caseworkers at ease. (Was the child handed off between siblings as a form of welfare fraud?)

They got lucky with the thyroid test in the Boston (Keegan) case. The chimeric cells can be anywhere, in any proportions. (There's no guarantee the complete thyroid, or complete ovary is a single genome... so the 'luck of the stick' could change the results.)

Siblings don't share a guaranteed 50% of DNA, there's actually a chance (though remote) that they'd share no DNA.

Assuming each chromosome of a pair has an equal chance of being passed on (someone else got info on what, if anything, is known to change these odds?). So there are 2 x 2^23 possible chromosome sets. While the odds are low, it's entirely possible that two siblings with the same parents would have less than 50% of their DNA in common.

In the case of a chimeric individual, if they already were below 50% DNA in common, then the children they produce might appear as only distantly related, nieces and nephews at best.

True in theory, but unlikely to the point of irrelevant.

Assuming chromosomes are inhereted whole, the probability that siblings share less than 25% DNA - the equivalent of one further generation away - is 0.00531 (0.5% chance).

But chromosomes are not inherited whole - they recombine. This hugely reduces the probability even from that start point.

Reply on news.ycombinator.com