Some DNA results are suggesting that the Wife of the John Cecill who came to St. Mary's County (SMC), Maryland in 1658 may have been Mary Mills, the daughter of one of his neighbors, Peter Mills. See the “Mary Dant Mystery ” section and the "Current Test Results" below.
We have obtained whole genome sequencing (WGS) results for four individuals. The testing was done by Dante Labs.
The effort looking at John Cecill’s possible wife is using the whole genome sequencing (WGS) data of 3 descendants of John Cecill [Dave, Jason, and Clemie] and 1 descendant of Peter Mills (1635, Holland-1687, SMC) [Peter].
Dave and Jason share a common ancestor with John’s grandson , James, Jr. (?-1785, SMC) thru John’s son James, Sr. (?-~1717, SMC). We believe that Clemie is descended from one of John’s other son’s (possibly his eldest son, John). In general these connections are back 9- to 10- generations or more This makes analysis difficult as the amount of DNA being passed down over those time periods is expected to be low.
The first approach was to attempt a Principle Component Analysis (PCA) on a data file with the DNA markers (or mutations) that had been found for Peter, Dave, Jason, Clemie, and another individual, HG00117. HG00117 participated in the 1000 genomes project and so his WGS test results were freely available. HG00117 is a male and had been born in Scotland or England. It is unlikely that he would be related to any of John Cecil’s descendants or to Peter. He was included with the others as a way to check the reasonableness of any results. This is a result from that PCA effort:
Reading along the bottom axis it is clear that Clemie, Peter, Jason and Dave are concentrated to the left side of the plot, while HG00117 is well separated from them at the right side of the plot. This means that HG00117 has no connection, or recent common ancestor, with the other people; just as we had assumed.
However, since Clemie, Peter, Jason and Dave are concentrated at the left side of the plot it seems likely that they are all actually related in some way.
Since this looked promising, We used another method, to look at the data to try to identify DNA segments that these people might share. For this case there are two types of possible DNA segments: identical by descent (IBD) meaning that the shared segments were passed down from a common ancestor (which we want to find) or identical by state (IBS) meaning that while the segments are identical, they just happen to be the same by chance and were not passed down from a common ancestor (we don’t want these segments).
To look for possible IBD segments we used a program called Beagle from the University of Washington, a DNA file handling program, bcftools, and some R script programs prepared for us by ChatGPT.
This is a simplified copy of the data that resulted from that effort:
This shows the IBD segments of DNA that the various individuals seem to share and which we assume were inherited from a common ancestor. The column marked CHR stands for Chromosome. The first record with Chromosome 12, for example, is saying that Clemie and Peter share a segment of DNA on Chromosome 12 that starts at location 34,849,374 and ends at location 37,857,375. This segment is 3,008,001 base pairs long. The LOD column provides an indication as to how reliable that segment is (a higher number is better).
The result of all of this is that it seems there is a good chance that Peter does share DNA with descendants of John Cecill. This implies that there is a good chance that Mary Mills may have been John Cecill’s wife.
Because we are trying to push the DNA results so far back in time, we still want to get verification of these results from a different approach. For this we are currently working to apply the Ascertained Sequentially Markovian Coalescent (ASMC) method developed at Oxford University. Those results will be provided when available.
(Updated 2025)