My last major Blog on Hartley STRs was about 2 and a half years ago. Things don’t always happen quickly in the realm of YDNA. However, as a result of my distant Hartley cousin Steve taking a BIg Y test, he now has new 111 STR results.
Steve, whose Big Y test is processing is now my closest Hartley STR match at the 111 STR level.
My Previous STR Tree
STR trees are difficult to draw and the results can be ambiguous. That is why the Big Y test which uses less ambiguous SNPs is a better test overall. Here is the tree I made over 2 and a half years ago:
Note that there were three Hartley’s who had taken the 111 STR test at the time. My guess is that one was too distant to be considered a match. I am Trawden Hartley in the diagram.
These people are also listed at the Hartley YDNA Project site:
I have highlighted the Hartley’s in my group who have tested to 111 STRs. The difference between the upper and lower group is that the administrator wanted the upper group to do a Big Y or more SNP testing.
- David Hartley 1797 – this is for Steve who is awaiting his Big Y results
- William Shepherd Hartley Lancashire 1851 – This is Wray Hartley in the diagram above
- Thomas Hartley about 1769 – This is the Thornton Hartley in the diagram above
- Robert Hartley – This is me, Trawden Hartley in the diagram above
In my 2015 Blog, I had this three person Hartley 111 STR signature:
In that signature, I tried to take account of the older SNPs to get the ancestral values. This time, I just took the simple mode of the four tests:
I was having trouble figuring out who Ross was. The reason for this is that he doesn’t appear to be in the Hartley YDNA project. I match the David ancestor above at a GD of 7. I match the other two at more than 9. The problem with doing it by hand is that it is easy to miss things. Fortunately, there is a way in Excel to choose differences.
Also notice that Excel interprets some of the values as dates, so care must be taken to format and copy and paste as text. Excel otherwise interprets 11-14 as 14-Nov.
Here is the new and improved version:
There were some discrepancies between what I did before I what I have now. I’ll go with my new chart for now:
Here I have a new Hartley A11132 Hartley 111 STR Mode. When there was a tie on the mode I used the higher value as I noticed that was what FTDNA did. It may be more accurate to consider the other testers for the first 67 STRs or go back to an earlier SNP as I did previously. For DYS447, I had a Hartley mode of 25 previously, so I’ll use that.
I’ll also change the marker before DYS447 which is DYS445. The older SNPs upstream of A11132 had a value of 11, so I’ll use that for the mode. This actually makes a big difference. DYS445 is a very slow marker changing at a rate of 0.16 per 1,000 generations. A male generation is 35.0 years. That means that this marker has a 16% chance of changing every 35,000 years or one chance of changing every 218,750 years! That tells me that the marker should be 11 because the chance of this changing to 12 for Hartley and then back again would take over 400,000 years. However, in another paper, I see a rate of:
I take that to be 2.16 per 1,000 generations. So that is a big difference. I also see this:
And I see that the 0.00216 was a mistake. At least I’m not the only one who makes mistakes.
Analyzing the Numbers
Here if a STR value went up, I gave it a pink. If it went down from the mode, I gave it a blue.
Here is some more information on rates:
I think the green, yellow and ornage color-coded numbers are old, but I like the colors as it gives a relative speed of mutation.
Rebuilding the A11132 Hartley STR Tree
The line that I have for the mode is considered to be the oldest value – though there is no guarantee – especially for the faster dark green markers. I added a column on the right for distance from the mode:
That means if I did it right, Steve at the top is closest to the mode or has the oldest combination of STRs. I am next with 4 differences from the mode. I’ll refigure myself with Steve:
Part of why I wanted Steve to take the Big Y test is because he seemed more closely related to me. I already saw that Steve and I shared the older value of 11 for DYS455. I think that is the major split for this group. Then I see that Steve and I also share the newer value of 26 for SYS447. Then after that Steve and I will split off from each other. Steve has the very slow moving newer marker of DYS435 and I have two other newer faster moving light green markers. I may ignore the darker green STRs for now as they could back-mutate more easily.
Here is the first cut:
Above is the Hartley Mode. I’m not sure if I displayed this the best way. At some point, an ancestor of William and Thomas had a mutation in marker 455 from 11 to 12. This split the A11132 Hartley line into two lines. Likewise, the assumption is that at some point, an ancestor of David and Robert had a mutation of 447 which went from 25 repeats to 26 repeats. Alternatively, an ancestor of David and an ancestor of Robert could have had parallel independent mutations. However, I think that this would be less likely. We don’t know which mutated first – 447 or 455 so I have them at equal levels.
Finally, everyone ended up on their own branch.
Fine-Tuning the STR Tree
Next, I can add in the fast markers. While doing that, I see one of my mutations that I missed:
Here is the new tree:
The other thing that I gather from this tree is that the common ancestor of David and Robert Hartley could be more recent than the ancestor of William and Thomas. That is because David and Robert have fewer mutations on their line. David and Robert have a total of 6 William and Thomas have a total of 10. I have from a previous Blog that STRs mutate at the 111 level on average every 125 years. I’m not sure if that is still a valid number.
I’ll take the average STR mutations for each branch, multiply by 125, then add about 50 years for average age (maybe low as I’m 63):
Assuming that was right, I’m not sure how to date the Hartley Mode.
Cross-Checking Dates Using Big Y
I currently match one other Hartley at FTDNA under A11132:
This shows that we have an average of 2 private variants between us. We used to have three, so that number went down for some reason. Perhaps the analysis was refined. The rule is that you can multiply this number by 144 years to get the years to common ancestor. That would be 288 plus my age of 63 or about 350 years ago. That means that it is possible that our common ancestor was as recent as 1670.
That is getting near the Big Y’s ancestor Samuel and/or Edward Hartley born 1666:
An interesting thing about this Big Y tester is that his DYS455 value is 11 and his DYS447 is 25. That brings about this unlikely scenario:
Here I have the other Big Y tester with 1666 ancestor on the left as his 455 is 11 and his 447 is 25. So that puts him above the common ancestor of David and Robert Hartley. That means that if I made the tree right, My common ancestor with Samuel Hartley could be around 1600, my common ancestor with the David Hartley Line could be around 1700 and the Hartley Mode could be around 1500. That would put the common ancestor of the William and Thomas line too early at 1345. That could possibly also be at 1600. Something to think about.
Here is my rough guess:
I think that the relative dating holds together somewhat. It still seems that the David and Robert Line (Steve and I) have the most recent common ancestor in this group.
Summary and Conclusions
- I played around with Steve’s new 111 STR results and created a possible STR tree
- The dating was interesting but it ran into a fairly recent proposed date based on Big Y testing. This Big Y dating could be more refined if the tester had uploaded his results to YFull. YFull has had a good reputation in the dating department.
- I was able to priortize some of the STRs based on their speed of change. The more slowly changing STRs should be the most important ones.
- I came up with some rough guesses on dating based on a lot of assumptions.
- The newer more recent Big Y tester’s common ancestor dating makes review of the other Big Y tester’s genealogy more relevant.
- A lot of this work is in anticipation of Steve’s upcoming Big Y results.
- The STRs are not as accurate as the SNPs produced by the Big Y, but they are interesting to play around with in making predictions. They can also be used in conjunction with the Big Y information.