Big boys don't recombine.

This is one of the most unnoticed and yet most interesting papers I've read in the last 10 years. Using a large amount of human genetic data, the authors calculated the recombination frequency as a function of position on all 22 somatic human chromosomes.

Brief 'splainer: most of us have 22 pairs of somatic chromosomes. One member of each pair comes from our father, one from our mother. However, in the process of making sperm and eggs, our father and mother took each pair of chromosomes they had, and did something called recombination; they spliced the chromosomes together, in one or two places on each chromosome, typically, to produce two new chromosomes, each partly granddad, partly grandma. Then during meiosis they put one of the chromosomes in a single developing sperm or egg. So if grandad's orginal chromosome was blue, and grandma's was red, you inherit one of these two chromosomes from dad (and similarly from mom).

X-chromosomes also recombine, but only in women, because men only have one. Y-chromosomes don't recombine ever.

What's interesting about the paper is that it looked at where the recombination events occur in both men and women. This just one example, chromosome 10. The red line is the frequency of recombination in women; the blue line (pointing downwards) the frequency in men.

There are always a lot of recombination events at the chromosome ends; that's because these are the telomeres, that have highly repetitive DNA sequences which make recombination more likely. There are also hot spots towards the middle of chromosomes, but what's remarkable is there are far more hot-spot recombination events in women than in men. Men, in fact, seem to recombine rather rarely, except at a few locations. So what does this mean? It means if you inherit DNA down the male line, where all the meiosis was in sperm cell progenitors, the DNA in the center of your chromosomes has remained in fairly large chunks. Down the female line, it's far more fragmented. This, IMO, is a tool hardly anyone is using, and it could give more detailed genealogical information.

I came across this by comparing the DNA between my wife and a very distant relative (6th cousin). There should be very little DNA shared, and it should be highly fragmented. But it isn't. It's in one huge chunk, on chromosome 10. And several other distant DNA relatives share the same chunk. I surmise they're all descended, mostly along the male line, from the same 17th century Montbéliardais gentleman. My wife is on the direct male line (very similar surname); others have one or two or three women mixed in their descent(so their surname is different), but the inheritance is predominantly male.

This totally screws up the genetic genealogy. My wife and her DNA relative are listed as probable fourth cousins; they are in fact two generations (4 steps) more distant, but only because they happened serendipitously to inherit this big piece of DNA. If the other chromosome of the pair had been passed on, anywhere down the line, they wouldn't register as DNA relatives at all.

So if you've done 23andme.com or familytreeDNA.com and gotten DNA matches, look at them. If they're all broken up into little bits, they're likely from your mother's side (and the other person's mother's side). If they're big and chunky, then they're mostly from the male side.