Home
This is a genealogy site focusing on the 'Rox2' yDNA cluster. The close similarity of y-STR markers shared between different individuals can indicate common paternal ancestry. Paternal yDNA is to genetic genealogy what the surname is to classical genealogy. Whereas a surname is liable to suddenly change for numerous reasons, the y chromosome is passed down virtually unchanged to all descendants from father to son, although small differences are inherited over time due to natural mutations that occasionally occur when the y chromosome is replicated. Those differences can be used to date haplogroups.
Rox2's 'signature' y-STR pattern was identified in 2005. With advances in hobbyist DNA testing technology and increasing numbers of people participating in DNA genealogy the initially small group of 25 and 37 STR marker kits is now a large SNP-defined haplogroup. Based on currently available data it appears that Rox2's Early Medieval paternal common ancestor produced at least eight 'sons', i.e. eight parallel yDNA lineages. The SNP resolution is not sufficient to be completely certain that they were actual brothers but the eight lineage founders were certainly very close paternal relations and they all appear at once as parallel individual branches on the phylogenetic tree. Due to prolific expansions in some of those lineages around 1000 years ago the STR haplotypes of Rox2's modern descendants are numerous enough and similar enough to be easily identified today - i.e. they match each other at certain 'key' STR markers in the 111 marker Family Tree DNA (FTDNA) test. SNP testing can then confirm that a STR match is a member of a particular subclade (a related branch, or subgroup).
TMRCA (Time To Most Recent Common Ancestor) estimates indicate that the Rox2 patriarch lived around 1300+ years ago (in my estimation he was born in around 700 AD plus or minus a margin of error). The margin of error means that an origin over 200 years earlier is not out of the question. Age estimates made with modern haplotypes are not precise, yet it is possible to confidently state that the Rox2 founder was probably alive in the Early Middle Ages (c. 400 AD to c. 1100 AD). The extremely shallow phylogenetic structure suggests that Rox2 is the result of an isolated founding event after a long genetic bottleneck - no branching has been detected in the c. 3000 years that have passed between the Bronze Age and Rox2's sudden expansion the Early Middle Ages. The eight 'brothers' produced parallel yDNA lineages, or subclades of their own - a 'star cluster'. Surnames only became fixed many centuries after the time of Rox2's founding and are liable to change in a generation. Prior to relatively recent emigration to worldwide colonies Rox2 had a northern European geographical distribution. The most southerly European Rox2 match so far is from northern France and the most northerly match comes from the north of Norway. The most easterly Rox2 match is from Finland and the most westerly is from Ireland.
Rox2's complex SNP phylogeny was discovered through NGS testing with Full Genomes Corporation (FGC ) in 2014. 'Rox2' is an early name coined by Jim Turner in 2005 - the term has no particular meaning (geographical, historical etc.) but it continues to be a very handy shorthand moniker or nickname for a haplogroup that is known by several different alphanumeric lead SNP names on several different phylogenetic trees. STRs remain useful for identification but NGS testing can now identify and accurately place a new Rox2 match on a phylogenetic tree, like the Big Tree and FTDNA's Block Tree.
ORIGINS
There is no firm evidence yet of where exactly Rox2's Early Medieval ancestor came from, that cannot be determined using modern yDNA alone. That he lived in northern Europe in the vicinity of the North Sea is a reasonable assumption given Rox2's age and modern distribution. However, hobbyist databases do not have uniform Europe-wide coverage and countries outside Ireland and the British Isles have a relatively low representation in those databases. 4000 year-old branch points in phylogenetic trees made using mainly North American present day results can provide clues but do not reveal early geographic origins with any certainty. The place of a haplogroup's highest modern frequency is not necessarily the place of ancient origin and, like water, the descendants of people might flow from a source and pool elsewhere. For in-depth discussion see the Ancient DF27 section and the Bronze Age, Iron Age and Medieval sub pages.
Studies of isotopic signatures from ancient burials have shown that there was great mobility across Europe. The relatively new science of archaeogentics is beginning to solve some of the many puzzles through radiocarbon dating and analysis of archaeological remains bearing uniparental markers in Europe. The places of birth of ancient yDNA subclades (mobile prehistoric men) might never be known for certain but archaeology and archaeogenetics can provide answers in some cases. The testing of ancient DNA can pin a sublade down to a particular place at a particular time. Studies in 2021 and 2022 have indicated that branches of DF27 were established in most of northern continental Europe by the Iron Age and had moved to different parts of northwest Europe in the Migration Period of the Early Middle Ages. The many different parallel lineages of DF27 will of course have quite different migration histories as they have been active across Europe since the Bronze Age.
The currently incongruous characteristics of Rox2 - i.e. DF27's lower frequency in western Britain and Ireland and Rox2's very long 3000+ year bottleneck followed by a sudden and relatively recent widespread demographic expansion by many parallel individual lineages in the Early Middle Ages - do nevertheless stand out in the available data. If the Rox2 founder's ancestors were active in Britain for over 3000 years, why did they leave no trace in the British Isles and Ireland until the Early Medieval period - in an area of Europe that currently has the highest hobbyist yDNA testing coverage in the world? L21>DF13 subclades by comparison generally have long yDNA lineages with many branch points stretching far back into prehistory in Britain and Ireland.
Significant population movement occurred between the time of the decline of the Western Roman Empire in about the early-fifth century and the time of the Norman Conquest in 1066 and into the twelfth century. Rox2 was probably born some time between those two events. A move from one country to another can explain a new subclade's appearance 'out of the blue' following a very long yDNA bottleneck but there can be other reasons. There appears to be a higher modern DF27 frequency on the continental coast of the southern North Sea basin than in other places in northern Europe. What is know with some degree of certainty at this stage is that the descendants of at least eight contemporary close paternal relations were living and having children around a wide area of the British Isles, Ireland and Scandinavia, in the Middle Ages. For the phylogeny and further discussion see the Founder page and the Geographical Distribution page.
DF27
This page shows early R1b-P312 branching. The ancestor of R1b-P312, that is R1b-L151, was from an early Corded Ware Culture lineage with origins in the east, on or near the forest steppe of Eastern Europe. There might be even earlier links to a Yamanaya-related culture of Western Steppe Herders (WSH), possibly in Sredny Stog culture. Prehistoric Corded Ware clans covered an extremely wide area. R1b-P312 was perhaps born just before the formation of Single Grave Culture in northwestern Europe c. 2850 BC. After a few generations Mr. P312's early descendants founded many of the paternal lineages of the enigmatic Late Neolithic and Early Bronze Age cultures of western Europe. The more prolific descendants of R1b-P312>Z40481>ZZ11, born after about 3000 BC + or - a margin of error, are DF27 and U152. Together they account for a large proportion of the the yDNA lineages in the European population. DF27, the subclade of P312 that Rox2 descends from, is one of the largest, most widespread but possibly least well-understood of the three big R1b-P312 subclades so far. DF27 was discovered four years after the more well-known branches L21 and U152 were first identified. Its SNPs are located in a difficult-to-read part of the y chromosome, the technology used in SNP tests has difficulty identifying DF27 in degraded archaeological samples. Based on modern hobbyist data DF27 appears to make up around one quarter of all the R1b-P312 subclades distributed across Europe but has a highest present-day frequency in France, Iberia, Netherlands and Belgium. In the west of Europe DF27's lowest frequency is in Ireland, Scotland and Wales. Within the British Isles its highest frequency is found in England. The place where the four largest R1b-L151 (aka P310) subclades (DF27, U152, Z290/L21 and U106) are found in roughly equal numbers today appears to be northwestern France (Normandy and Centre-val de Loire).
After more extensive post-2014 DF27 NGS testing (using FGC and FTDNA's Big Y) and analysis by Alex Williamson at the Big Tree two large ancient groups just below DF27 were revealed. They are ZZ12 and the smaller parallel branch that had already been known for several years, Z195/Z196. Two further early branches were revealed in 2018. They are BY168384 (hg38 position 19555478 T>C), and FT273897. The more recently discovered deep subclades currently have few members. More early branches might turn up in the future. According to FTDNA Haplotree (January 2022) two more 'brothers' of ZZ11 exist downstream of P312>Z46516, i.e. FGC84729 and BY189791. Also ZZ11 has another 'son', Z140446 (the new 'brother' of U152 and DF27).
DF27>ZZ12 stands out within R1b-P312 having appeared to have produced a big founding event in the Late Neolithic/Early Bronze Age. The phylogenetic trees indicate that a large cluster of 'brothers' suddenly appear in about the Late Neolithic/beginning of the Bronze Age - brothers who went on to found yDNA lineages of their own that survive to this day. FTDNA Discover (November 2022) indicates 34 branches. The scale of this prehistoric founding event just below DF27>ZZ12 dwarfs Rox2's Early Medieval founding event more than 3000 years later.
Rox2 is DF27>ZZ12+. Its ancestral branch immediately below ZZ12 is Z46512 (hg38 4365426 T>AC )/FTT1, then FGC78762 (hg38 10777194 G>T ). Next is the big 'son' ZZ19/ZZ20 (hg19 25938772 T>A, hg38 22224478 A>G).
Subclades under ZZ19 are widespread and descendants are found all across Europe, including in Sweden, Finland, Poland, Denmark, Czech Republic, Germany, Austria, Armenia, Italy, France, Spain, Portugal, Britain and Ireland. ZZ19 is a large group, making up over one third of the total of all DF27>ZZ12 on the Big Tree. FTDNA's 'Discover (Beta)' feature reveals that Babe Ruth belonged to ZZ19 (Z31644 branch). See the Ruth DNA Project. Ruth's paternal grandparent is said to trace back to Prussia in Germany. At the FTDNA Haplotree (counted 4th June 2022) ZZ19 made up 21% of all DF27 participants and 39% of DF27>ZZ12. Rox2 makes up 14% of ZZ19 participants at the FTDNA Haplotree.
The first ancient individual with SNPs downstream of DF27>ZZ12>ZZ19 to be identified in a study was VK261 from Population genomics of the Viking world, Ashot Margaryan et al., 2020. VK261 (ZZ19>Z31644>FGC13128>FGC78763>BY64643) was an executed Viking (dated 10th-11th centuries AD) from the Ridgeway Hill Mass Grave, Dorset, England. Further examples of continental northern European-origin ZZ19 are also found in the era preceding the Viking Period, in the Migration Period, in The Anglo-Saxon migration and the formation of the early English gene pool, Gretzinger et al., 2022. See the Medieval DF27 page. ZZ19 burials were identified in Large-scale migration into Britain during the Middle to Late Bronze Age, Nick Patterson et al., 2021, Nature. Many of the ZZ19 results are from Iron Age burials in East Yorkshire, England and belong to the distinctive local Arras culture. See the Iron Age DF27 page. The earliest ZZ19 to be discovered so far looks to be I15033 from Bronze Age Amorica, at Port Blanc, Quiberon, Morbihan, Brittany, France. 1891–1743 cal BCE (c. 1817 BC). I15033 was listed as P312 in the Patterson et al. paper but Alex Williamson found the sample to be ZZ19.
Below ZZ19 are two 'brothers' Z34609 and Z31644. These ancient branches probably existed c. 2500 BC at a time when Bell Beaker people were expanding across Europe. Z31644 was prolific and had at least fifteen 'sons' (FTDNA Haplotree, June 2022) and 'brother' Z34609, from whom Rox2 descends, had at least six. The presence of Z34609* and Z31644* kits in modern phylogenetic trees indicate that more parallel subclades await discovery.
'Son' of ZZ19>Z34609, Z2571, has two ancient parallel subclades descending from it, they are FGC11380 and CTS11567/FT4427. Rox2 makes up 70% of Z2571 participants on the FTDNA Haplotree (June 2022).
ROX2 SNPS
Z2571 (hg19 23076115 C>G, hg38 20914229 C>G) is a deep DF27 SNP. ISOGG: R1b1a1b1a1a2a6. Formed c. 2500 BC +/- margin of error (YFull).
FGC11380 (hg19 23098886 T>C, hg38 20937000 T>C). Early branch point (with equivalent FGC11385 position 14422221 G>T/12301496 G>T on FTDNA Block Tree). Big Tree FGC11380 Y8841, 22220783-C-T, YFull's R-Y8841: FGC11385 * FGC11384 * FGC11380/Y8841. Formed c. 2500 BC, TMRCA c. 2200 BC +/- margin of error (YFull).
FGC11397 (hg19 7332619 G>A, hg38 7464578 G>A). 'Rox2'. A currently phylogenetically equivalent SNP, one of a block of dozens of equivalent SNPs below FGC11380/Y8842 that can help define Rox2. It is not known know which of these SNPs came first or last. The ancient R1b-M269 clade that R1b-P312 eventually descended from has many more equivalent SNPs, i.e. 108, and took thousands of years to form. The Big Tree leads with FGC11388. YFull lead with Y8397 (FGC name FGC11374). Another possibly early equivalent is FGC11381 (hg19 6720487 C>T/hg38 6852446 C>T). The FTDNA Haplotree occasionally uses FGC11395/Y8842 (position hg19 14861066 C>A/hg38 12749132 C>A), FGC11411/Y8406 (hg19 19116326 T>C/hg38 17004446 T>C ) and BY741/Y8398 (hg19 7907577 C>G, hg38 8039536 C>G), currently at the same level as FGC11397, as Rox2's lead SNPs.
TESTING FOR ROX2
The simplest yDNA testing strategy for clear identification of a Rox2 match is to get a NGS SNP test (e.g. Big Y, YSEQ or FGC). If happy sharing results and comparing with others make sure privacy settings are set to allow that. NGS SNP results can also be added for free to Big Tree, a public phylogenetic tree. New entrants to the hobby should get a 111 STR test and think about getting a Big Y test. 111 STRs have been, and continue to be, an invaluable tool in the understanding of Rox2. There are 37, 67 and 111 marker Rox2 matches across different FTDNA surname projects who are unaware of SNPs or the fact that they match Rox2. Close STR and Big Y matches can be contacted via email from the FTDNA website.
All Rox2 matches share a large number of relatively recent SNPs in common with each other across their respective y chromosomes. Those SNPs are presented by the testing lab as a block of SNPs on the phylogentic trees. That large phylogenetically equivalent SNP block shared by all Rox2 matches might be imagined as a father-to-son chain of descent with dozens of SNP 'links' in that chain below Z2571>FGC11380. However, we don't know the chronological order of those links. A branch in the large shared block might turn up in the future and could potentially give some clues about the origin of Rox2. The very long SNP 'bottleneck' ends some time around 400-700 AD when several parallel branches appear to suddenly expand all at once - in a 'founding event'.
'SONS' OF ROX2
There are currently eight named branches. A17453, FGC11414 and FT171815 look to have been responsible for large early founding events of their own - they each produced a similar number of 'sons' as their Rox2 father. In addition there is a FGC11397* paragroup at that might be imagined as occasionally being a holding group for analysis by FTDNA. I try to keep the information updated but new Big Y kits are regularly added to the FTDNA database and branch names at all levels on the phylogenetic tree can change.
Note: After initial Big Y-700 results come in it can take time for them to 'settle down' into their final positions on the FTDNA Block Tree following analysis of the aggregate data at FTDNA.
The NGS-identified and 'named' Rox2 deep subclades on the Big Tree and Block Tree (see FTDNA Haplotree and Discover) are:
FGC11414 had a large founding event very soon after his father (FGC11397/Mr. Rox2) and has several branches below or 'downstream' of it - they are parallel 'sons' of Mr. FGC11414. Four branches have been named so far on FTDNA's Block Tree: FGC11419, BY21590, BY21591, Y69563. Some new kits have not uploaded results to the Big Tree yet and as a result FGC11414 coverage is inaccurate there. A subclade with members from northern Sweden, Norway and Finland is represented by FGC11414>BY21590. A medieval founder currently looks to have had two 'sons', BY43192 and FT249214, c. 1300 AD plus or minus a margin of error (as seen on Block Tree, not Big Tree). The FT249214 branch (plus six more phylogenetically equivalent SNPs) currently contains modern descendants in northern Norway and Sweden. A descendant of the other branch, BY43192, had four sons in the mid-1500s AD in northern Sweden and they became the ancestors of the majority of the currently known Swedish Rox2 kits. No BY21590 has been found outside Scandinavia yet. See this page for more STR and SNP-based age estimates and Bo-Göran Persson's Nordic Rox2 Y-DNA page for more details. Another FGC11414 'son', BY21591, contains Big Y kits representing old West Riding of Yorkshire families. FTDNA's Haplotree shows two kits who are BY21591* and a named branch, BY49975. Big Tree leads this branch with BY49857. See the FTDNA Robertshaw DNA Project. BY21591 (just one SNP, no equivalents) is a 'son' of FGC11414 and was born at a very early point in Rox2's history, around 1250 years ago +/- a generous margin of error. The proximity of two separate, parallel early BY21591* ancient 'brother' lineages whose currently-known representatives respectively bear the names Smithson and Falkiner today, with long pedigrees in broadly the same region that BY21591>BY49857 is from, is interesting and suggests that BY21591 may have ancient continuity in that area of Yorkshire. As with all subclades immediately below Rox2, the places of birth of the subclade founders' 'sons' could be nearby, or distant from, the location that their father was in nine months previously.
A17453 - notes regarding A16311 (18502051-G-T hg38), below A17453 on the Big Tree, say '*Mutations whose exact position can't be determined precisely from NGS tests, such as those in palindromes, are shown with a pink background.' FTDNA's Haplotree/Block Tree doesn't have A16311 but has FT289633 as an equivalent to A17453. Mr. A17453 appears to have produced a very large Early Medieval founding event, like his Rox2 'father' and FGC11414 'brother'. There were possibly at least nine 'sons' of A17453. Block Tree order: 1. BY21592/Y33105, 2. BY21588, 3. BY21584/A16155, 4. BY145620, 5. PH1941 (1000 Genomes kit?), plus three A17453* kits. There are many more matches to this subclade's 'signature' STR pattern in the FTDNA database who will go on to find they match A17453 should they become aware of Big Y/NGS SNP testing and the Big Tree. It is always a good idea to email close 67 and 111 marker STR matches, let them know about Rox2, Big Y/SNPs and the Big Tree.
Y17787/FT14357 (YFull use equivalent Y17484 as lead SNP). Big Tree has Y17484 as an equivalent of FT14357 but FTDNA Block Tree have FT14357 above Y17787 and now lead with that SNP. The new Big Y kits that led to the change in early branching in this subclade are missing from Big Tree but have formed a very early (Early Middle Ages, possibly eighth century AD) branch. It is called R-FT253057 (plus one equivalent FT253478) on FTDNA's Block Tree. A branch point in that lineage, FT253663, forms in around the tenth century and two branches have had parallel existence for about 1000 years. Mr. FT14357, a 'son' of Mr. Rox2 himself, was the father of FT253057 and Y17787. Differences between the two phylogenetic trees would be resolved if the FT253057 kits shared their results on the Big Tree. Many Macauley families below Y17787 possibly stem from a founding event that produced at least two 'sons' (Y17484 and Y17787*) in around the mid-thirteenth century, probably in the northwest of the Isles. There are currently five known early sons of Y17484. They are represented by Y78091, Y20979, Y30689 and Y83213 and one that will remain Y17484* until joined by a fellow branch member. The FTDNA MacAuley DNA project is here: link
BY21579/YSEQ's A12541 leads a phylogenetically equivalent SNP block on The Big Tree and Block Tree. With the inclusion of Big Y-700 Cameron kit (November 2020) the early block shows a branch point, possibly in the mid- to late-eleventh century AD +/- a margin of error, when two 'sons' of BY21579 appear. They are BY21578 and the BY21579* Cameron kit with origins in Scotland who will discover more branching when another more closely matching person takes a Big Y test in the future. The two lines spend many generations in parallel until Mr. BY21578 had five sons in around the 1500s AD. His descendants include the Irish Hickey/Higgins family and several other families that share the same paternal ancestry leading back to that sixteenth century founder. One son is BY30718, plus one equivalent SNP, the other four remain parallel BY21578* lines for now. After a few more generations three sons of Mr. BY21518 are born in around the mid seventeenth century (+/- a margin of error). See the FTDNA Hickey DNA project and the FTDNA Cameron DNA project (BY21579*).
FGC39763 leads a block of 3 phylogenetic equivalents on Block Tree. A branch (FGC39762) forms when two 'brothers' are born in the early-tenth century. Below that are two long equivalent SNP blocks, one containing Big Y kits from relatively recently related Simpson men and the other with one Big Y-700 kit from northern France that remains FGC39763* for now and is only visible on the FTDNA Block Tree. Those two lineages have been existing in parallel for over 1000 years.
BY166898 (plus Big Tree equivalents Y2:13914-C-T [not on FTDNA Haplotree], BY168115, BY166898, BY168311 and 5413822 G>C/FT33483. FTDNA have BY168115 as the lead name of the Block Tree branch known as BY166898/13914-C-T at the Big Tree. SNP counting currently suggest that the common ancestor of BY166504 and FT64296 may have lived in around the early 1100s AD (+/- a margin of error). The FT64296 Scott families have documented origins that trace back to a man who emigrated from Perthshire in the late-seventeenth century. A founder looks to have had at least three 'sons' at about that time, FT63817 and two who remain FT64296*. N3461 Corner (me, North Yorkshire), is currently BY166504*. Big Tree includes a SNP that FTDNA omits from their list (K498, position 10213487 G>A , hg 38). BY194976 and equivalent BY195508 are in both Block Tree and Big Tree. BY166504* and BY194976 may have branched from their BY166504 common ancestor in broadly around the mid- to-late 1300s AD (+/- margin of error) and have existed in parallel since then.
FT171815 is the single SNP that represents this subclade and is a 'son' of Mr. Rox2. The downstream branches, i.e. the 'sons' of Mr. FT171815, share a male common ancestor who lived around 1250+ years ago at the very start of Rox2's existence. The Early Medieval 'brothers' are FT30548, FT199028, FT412827, FT251688 and one who, for now, remains FT171815*. Two of the yDNA branches below FT171815 represented by present-day Ker and Newton kits have origins in Scotland and Derbyshire respectively and have been in living in parallel since the Early Middle Ages. Mr. FT171815 looks to have produced a founding event similar to the one produced by his father, Mr. Rox2. Further Big Y testing by fellow matches from the five parallel branches will define them in more detail in the future and help name the currently FT171815* branch. Only the two kits mentioned above are on Big Tree. This branch appears to have a higher than average genetic difference from the Rox2 111 STR modal haplotype.
BY170312/FT132374 (Big Tree). A branch, BY169396 (BY169202 on Big Tree) forms in about the mid-tenth century.
The Block Tree differs from the Big Tree and branches can change after more in-depth analysis of the complex files at FTDNA. FTDNA's Block Tree shows that kit B351108 belongs to the FT171815 branch, so this means that there are currently eight known 'sons' of Rox2. However, at FTDNA's Block Tree there are over one dozen more unnamed and unidentifiable Rox2*/FGC11397* kits from several countries. Most of those will probably be kits that have only tested for the FGC11397 SNP itself, not Big Y, but some of those kits might belong to a yet-to-be-named branch. When/if matching Big Y kits turn up in the future, more named branches below Rox2 could be formed.
There are many downstream 'family' subclades and branches, so apologies if details about your branch are missing or are not covered in much depth. Some of the early subclades have phylogeny as complex as Rox2 itself and ideally require separate projects of their own. If you would like details of your branch to be included here consider creating a web page or blog and/or send a link to your FTDNA project.
The kits on the Big Tree and Block Tree represent only part of all Rox2 matches in the databases, many have STR results only. Two close Rox2 STR matches might occasionally have very low genetic distance from each other by coincidence at lower resolution. A 'lack of divergence' between two 67 marker haplotypes can occasionally give the incorrect impression that the two matching kits are closely related. Fortunately, NGS tests using SNPs can highlight these coincidental STR matches. For example, a member of one Rox2 subclade, BY21578, was an almost identical STR match (66/67 markers) to a member of parallel 'brother' Rox2 branch, Y17787. Big Y testing followed by placement on the Big Tree revealed that they are unrelated for over 1000 years.
