Ann Turner has written a utility for calculating the MRCA.
"The mutation rate is critical for statistical estimates of when the Most
Recent Common Ancestor (MRCA) of two people lived. The NUMBER of markers also
enters into the calculations. If two people match on all 25 out of 25
markers, that narrows the range of time when the MRCA probably lived,
compared to a match on 12 out of 12 markers. ..... I ... wrote a calculator so you can play with the numbers:
MRCA Calculator " (Edited from a message posted on the Genealogy-DNA List From: [email protected] (Ann Turner); Subject: Re: [DNA] Marker Choice Comparison; Date: Tue, 2 Apr 2002 22:07:29 EST).
When you use the MRCA calculator http://members.aol.com/dnacousins/MRCA.exe,
the results are only calculated for up to two mutations. Any more mutations
than that takes you too far into the past to be genealogically interesting.(Edited from a message posted on the Genealogy-DNA List From: [email protected] (Ann Turner); Sent: Monday, April 14, 2003 10:53 AM;
Subject: Re: [DNA] mutuation rate chart)
The Mutation Calculator is intended for
situations where you know the most recent common ancestor (or you want to do
a "what-if" calculation.) In (a) case, you could say, what if "A" and "B"
had a common ancestor in the very next generation after the paper trail runs
out?
You would have (the cumulative total number of generations between "A" and "B" and the MRCA as the number of) transmission events and enter the typical .002 mutation
rate. What's the probability that we would observe 0, 1, 2, 3 or more
mutations? Based on the data ... (you can form an opinion as to the number of mutations). ...
...Each new mutation occurs completely at random,
independent of any other mutations which have already occurred. The (number of)
mutations can be distributed in any way between the two lines.(Edited from a message posted on the Genealogy-DNA List From: [email protected] (Ann Turner); Sent: Monday, April 14, 2003 10:53 AM;
Subject: Re: [DNA] mutuation rate chart)
In the end, the thing you're really looking for from the DNA test is the simple indication that "yes, you
are related" or "no, you are not related". The more markers you test, the clearer that indication is likely to be, but you would need at least hundreds of markers to get beyond the bare yes-vs-no level. Further, the use of statistics to estimate the probability of a common ancestor becomes less rewarding as you increase the number of markers studied, so there are limitations on the theoretical usefulness of statistics in estimating the number of generations to a Most Recent Common Ancestor.