To start with, my brother in law's father wants to be anonymous, so it is necessary to change his name. He is henceforth Theophilus McKinstry.
Here is the haplogroup tree Family Tree DNA provided for Theophilus. On the first page, all of the major haplogroups are displayed and branching lines show how the haplogroups are related to each other. They are shown from the top, and the circle with the dot in it at the top is Genetic Adam who lived in Africa 60,000 years or more ago. (Genetic Adam is an anthropological construct and not Biblical Adam. Many other men were alive when Genetic Adam lived, but only one man has descendants living today. )
For simplicity, Family Tree DNA lists only one SNP mutation for each haplogroup, even though usually there are a number of SNP mutations for the haplogroup. You follow the heavy line from the upper left, and there are marks across one’s own clade.
Follow the thick heavy line to M223, I2b1. Under this are several SNP’s (not all shown), including M284, I2b1a. Under M284 are two daughter SNP’s (that so far always occur together); L126 and L137; I2b1a1..
Theophilus’s Family Tree DNA Haplotree
Here the blue carat’s ^ tell us that Family Tree DNA did not run a “backbone” SNP test to confirm I2b1. Family Tree DNA think they can be very confident that this is the M223 clade. They used a haplogroup predictor like this one that I used online, which predicts the Scottish subclade of M284 with near absolute certainty. I entered the STR markers into the haplogroup predictor using the dropdown boxes, and then clicked buttons to get the results.
The last two SNP's are colored in green because I tested Theophilus's Y DNA for the SNP's L126 and L37, and Theophilus has those mutations. This confirms that he belongs to the I2b1a1 clade. If there had been more mutations that one might test for, they'd have been colored in orange. .
However, once testing found the SNP’s L126 and L137, Family Tree DNA changed those SNP’s to green, meaning that they were found by SNP testing. They then changed I2b1a1 in every place where it is reported from I2b1 in red, meaning that it was predicted, to I2b1a1 in green, meaning that it was verified by SNP testing.
If I had had no idea what SNP's Theophilus might carry, I could have simply ordered a deep clade test. However, I entered the STR values into an online haplogroup and subclade predictor. These predictors make probalistic predictions of one's haplogroup and some subclades, based on one's STR markers. They predicted I2b1a Isles Scottish haplotype with near perfect confidence. In fact, the experts advised me I didn't really need to confirm the SNP's.
DYS 425 has three stars to signify that Family Tree DNA verified its null value several times. But M284+ people always have a null value at DYS 425. This means that around the time that this mutation happened, everyone who lived to pass it on to their descendants, was also descended from someone who lost all the repeats of this marker. I'm not sure why DYS 464 has two stars. Nothing is odd about its values.
There were two steps to the haplogroup predictor; one step to get the major haplogroup, and one step to get the subclade. Theophilus was lucky to have a subclade predicted with near perfect confidence. In R1b1b2 it is common to get several clades predicted with probabilities of being right no higher than .70, and often much lower. Very often only SNP testing can determine a subclade with reasonable accuracy.
Theophilus McKinstry’s Y DNA STR values entered into an online haplogroup predictor
I-S24 is P219, which in an apparently little used classification scheme is a daughter clade of M223 (I2b1), and a parent clade of M284 (I2b1a). => 100% means that the haplogroup predictor is absolutely certain this prediction is correct. By the rules the programmer programmed it with, it is very unlikely that this haplotype is not really M284.
The haplogroup predictor does not mention L126 and L137. This version of the haplogroup predictor was created before these two SNP's were discovered a little more than a year or two. It is now known that the Isles Scottish haplotype is a subgroup of L126 and L137.
Because new findings with Y DNA SNP mutations keep changing the haplogroup tree and the letters and numbers of its subclades, it is getting more common to call a subclade either by its major haplogroup and the SNP of its subclade, like I-M284, or simply by the SNP that respresents the subclade, like M284. This is particularly an issue with R2b1b2, which as a matter of fact is now (I think) R2b1a1, because the tree changes so rapidly and radically that the names of the clades have no meaning. The names of I2b1 clades will eventually change, because researchers are working very hard at finding more SNP mutations to break up the huge clade, or atleast to distinguish between very large and very old STR clusters like "Roots" and "Continental" (neither of which defines M284 which is a Neolithic or Bronze Age era clade of its own).