On Genetics and Genealogy, Part I
by James D. Shoemaker, M.D., Ph.D
As some of you may know, I am a physician working in the field of genetic diseases at St. Louis University. As I have become more familiar with the methods of family history, I have been struck by the many overlapping issues between the fields of genetics and genealogy. I would like to use this column to begin to describe some of them. First, just how related are we? Can modern molecular biology distinguish the fact that many of us are 8th , 9th or 10th cousins? Are two of us who might be Brewster cousins, for example, really any more related than any other two ethnically similar people? The answer is complicated, but an estimate may be made from the frequency of genetic diseases in our population. Among American Caucasians, the rate of diseases which result from inheriting the same mutated gene from both of your parents (a monogenic disease) has been estimated to be about 1% of all births1. This means that there is a fairly high background of inbreeding in our population. To illustrate this, we all know that most states forbid, for public health reasons, the marriage of first cousins. The risk of having a child with a monogenic disease for first cousins can be calculated and is about 6%, for second cousins 1.6%, and third cousins 0.8%2. Thus the general population, marrying at random, has about the same "inbreeding" risk as second- to-third cousins. Another way to think of this, more familiar to genealogists, is to estimate how many ancestors we all have in common. If you have 4 grandparents, 8 great-grandparents, etc., back x generations you have 2x grandparents. How many generations back do you have to go before you have a number of grandparents equal to the total population at the time? The population of the world 1000 years ago has been estimated at 250,000,000, and the population of Western Europe about one-tenth of that, 225 equals over 33 million, so if a generation is 20-40 years, only 25 generations ago, if there were no inbreeding, your ancestors would be every single Caucasian who has descendants. It is no exaggeration to say that it is likely that every person of Northern European descent is a descendant of Charlemagne, and assuming they had descendants who survive to this day, Joe the Baker and his wife, Kitty. Likewise, if Joe and Kitty lived 1000 years ago, and had only two kids, who each had two kids, who in turn over the centuries have had only two kids in orderly 30-year generations, the entire population today would be their descendants (actually over 8.5 billion descendants). When you average out all these effects, we all turn out to be about as similar as third cousins. This is not to say we are not all unique. We all have two copies each of about 80,000 genes, and for each one of us, probably about 20% of those genes have one copy that is "mutant" 3. Another way of looking at this is that one letter out of about 500 in our genetic code is probably a mutation, for a total of about 160,000 mutations per person4. Most of these mutations make no difference at all, but are useful for DNA identification techniques. Current molecular biological methods can follow patterns of these mutations through families and determine descent with a high degree of certainty. Incidently, according to experts in the field, paternity does not match what would be surmised from civil records in about 10% of cases. Yet for all their power, no DNA identification scheme today can deduce the most common physical traits of the person whose DNA is being analyzed--not their height, weight, skin color, hair or eye color. In the next Missouri Compact: how your maternal (mitochondrial) or paternal (Y chromosome) ancestry could be proven, once and for all.
genealogyinstlouis.accessgenealogy.com
* * * |
