McGrew Family DNA Project - Results

March - 2007 When the McGrew DNA Project began in February 2006, there were three goals in mind. The first was simply to see if it was possible to establish a genetic ancestry for Robert McGrew. The second was to try to establish a genetic link between Robert's northern McGrew's and the southern branches of the family. The third was to determine if McGrew's are related to the McGregor Clan of Scotland as my family lore suggested or, if not, to establish what Scots Clan we are related to.

Although we end the year having only eight Project members we have accomplished a good portion of these goals. When we found a 12/12 marker match (see comments on matches below) between kits 53756 (row 6) a known descendant of Robert McGrew and kit 52842 (row 7), a probable descendant of Robert who can't confirm one link, goal #1 was accomplished. It was very rewarding to have this occur with the first two sets of results processed. Kit 61737 (row 5) with an 11/12 match with kit 53756 (row 6) and a 35/37 match with kit 52842 (row 7) is also likely to be a descendant of Robert.

Our second goal was also achieved when a 12/12 match was found between kit 53756 (6) and kit 60818 (8), a known descendant of James A. McGrew, a southern McGrew. Kit 60818 (8) also had a 24/25 match with kit 52842 (7, me). Both James and Robert are believed to have immigrated to the Colonies from Omagh, N. Ireland. These lines will be named the "Omagh McGrew's".

In March, 2007, kit 53756 (row 6) upgraded from 12 markers to 37. Also in March, 2007, a second known descendant of Robert joined the Project (row 9). Our two known descendants match 22/25. However Row 9 matches row 8 23/25, row 7 24/25, and row 5 22/25. Some of these matches show less genetic distance for those individuals separated by more generations. Three markers are responsible for nearly all of the nonmatches. Our family may simply have a higher mutation rate than average for these markers. This level of match remains indicative of a recent genetic relationship among these individuals. Particularly interesting is the close match (24/25) between the known descendant of Robert (row 6) and the descendant of James A. (row 8), a southern McGrew.

Our third goal has been partially achieved. A typical set of McGregor DNA results found on their website have been examined and results suggest that there is no genetic relationship between McGregor's and McGrew's. Also over 1,000 DNA results from a Scottish Clans DNA database have been examined and no consistent matches have been found between our results and results within any clan. Some clans of interest (Fraser, Buchanan and Drummond) have very few published samples, and results from these clans will continue to be evaluated for matches.

There are several aspects of our family DNA results that are very interesting. First our heritage is Irish. We may be Scots, but we came to Scotland from Ireland. This fact became apparent as our DNA results were compared with others using a geographic analysis. Our DNA is very close to that of about 20% of Northern Ireland males. Many Scots in western lowland Scotland came to Scotland from Ireland, and there was an Irish kingdom, Dal Riata, that encompassed northern Ireland and western Scotland from about 500-900 A. D. Also the Roman word for the Irish was Scoti. Other immigrants to Scotland were the Picts, Angles, Norse and Jutes each with distinct DNA types. We are Irish.

Genetic genealogist David Wilson has proposed an R1b1c7 haplotype that is common in western Scotland and northern Ireland (see Seven of our results (rows 3-9) match this type. This even more firmly establishes us as being Irish before we were Scottish. Wilson wonders if we were among the earliest immigrants to Ireland possibly as far back as 8,000 B. C. Several suggested readings below discuss the origins of the British peoples.

Kits 65862 (4) and kit 66469 (3) are also interesting. They have an 11/12 marker match. This level of match suggests a possible genetic relationship (see below). Kit 66469 has not provided any known McGrew ancestors. Kit 65862 can trace his line back to William Van Buren McGrew born in KY about 1843. A possible common ancestor is John H. McGrew who had sons John L. and William V. John L. may be an ancestor of kit 66439. Particularly interesting is the possibility that this ancestor (John H. McGrew) is descended from Thomas McGrew who may be a brother or cousin of John Flood McGrew, a noted southern McGrew. If so we have another southern line that shows a possible genetic relationship to other northern and southern lines. We would need at least a 25 marker result set to more firmly establish the relationship between these individuals, and the existence of a genetic relationship with the Omagh McGrew's.

Kit 55457(2) shows no genetic relationship to any other Project members. The most distant ancestor reported is Alexander MacGruar who was born in Scotland in 1706 and immigrated to the Georgia/South Carolina area about 1735. MacGruar apparently immigrated directly from Scotland unlike Robert and James McGrew who were Scots-Irish and immigrated from Omagh, N. Ireland.

Kit 60717 which had traced its ancestry to Robert McGrew has a broken genetic link in its male line.

There have been questions from our Project members and others interested in our results about the number of matches required to suggest a genetic relationship between individuals. Differences in marker values between a father and son are the result of a mutation during the copying of the son's Y-DNA (transcription). There is an average error rate of 0.2-0.4% in the transcription process for each marker. The variation in error rates depends on both the identity of the marker and the individual doing the copying. Different markers have different mutation rates and different individuals (probably family lines since the transcription process is also inherited in the gene) have different mutation rates. If we initially assume a 0.2% error rate per marker copied and continue the process over 10 generations (the approximate genetic distance between us and Robert McGrew) we get a 2% error rate per marker. For 25 markers over 10
generations the total error rate would be about 40%. Thus for a 25 marker test 2 individuals separated by 10 generations from Robert McGrew should expect slightly less than a 1/2 chance that one marker has mutated. For a 0.4% rate the chance would be about 2/3. Thus we would expect at worst a 24/25 match. For 12 markers we would expect 12/12 with 11/12 suggesting a possible match particularly if the marker in question had a higher than average mutation rate (colored in red on the Family Tree chart).

Where do we go in 2007? We need more Project members. The biggest surprise has been the reluctance of McGrew's to join. We need a known descendant of Thomas McGrew. We need a descendant of John Flood McGrew to join. I have contacted several but, so far, to no avail. We need a descendant of Robert through other than his son James.

Suggested readings:

Oppenheimer, Stephen," The Origins of the British", Carroll & Graf Publishers, 2006.
Sykes, Bryan, "Saxons, Vikings, and Celts", W. W. Norton, 2006.
Smolenyak, Megan and Turner, Ann, "Trace Your Roots with DNA", Rodale, 2004.