Y Haplogroup L Resource page

Distribution Haplogroup L Y-DNA

By Crates (Own work) [GFDL (http://www.gnu.org/copyleft/fdl.html) or CC BY-SA 4.0-3.0-2.5-2.0-1.0 (http://creativecommons.org/licenses/by-sa/4.0-3.0-2.5-2.0-1.0)], via Wikimedia Commons.  Unmodified.

Introduction - Y-DNA, Haplogroups, SNPs, Haplotypes

The Y chromosome, and it's Y-DNA, are copied from father to son, down a strictly paternal lineage.  If I were to trace my entire direct ancestry back, I have two parents, four grandparents, eight great grandparents, sixteen great great grandparents.  Yet out of those sixteen great great grandparents (generation 5), who were born only circa 160 years ago, only one carried the Y-DNA that was passed down to me.  My eight great great grandmothers did not not inherit a Y chromosome from their fathers.  Most likely, my other seven great great grandfathers, carried distinctive differently marked Y-DNA.  Yet all sixteen biological great great grandparents have contributed to my overall atDNA (autosomal DNA).  Only one gave me my Y-DNA.  So you can see that Y-DNA represents only one narrow lineage.

Y-DNA, may on the face of it, appear to offer a limited understanding of total biological ancestry.  All sixteen of my great great grandparents were direct ancestors, not just the Y great great grandfather.  However, this lineage offers us evidence that can be genetically tracked, then mapped into relationship.  It could be done to ascertain parental, or non parental events.  It can be used to check the biological validity of relationship to cousins.  As more people investigate and record their haplogroups, haplotypes, STR markers, and SNPs, so we can for example, start to use them to map biological relationship further back.  Y-DNA is particularly useful, not only because of it's markers, but also because it can be plotted to surname studies.  In Western societies, the surname often follows the Y lineage for several generations.

However, Y-DNA (nor the maternal mtDNA) evidence doesn't just stop there.  As more people investigate, submit, and record their data from around the World - and as anthropologists and archaeologists add ancient DNA data from ancient and provenanced human remains to that record, so we can build and plot a world map of the human family, how it relates, how it was distributed globally throughout prehistory.

Both Y and mt DNA carries mutation markers, that define a HaplogroupA haplogroup is a family of shared descent.  These haplogroups are ancient.  The paternal Y-DNA haplogroup that this resource page is dedicated to has been designated as L.

However, mutations do not stop with the formation of a new haplogroup.  They continue through the generations.  As lineages divide between different sons, across many generations, so these mutations in the Y-DNA for example, continue to accumulate down the diverging lineages that once shared common descent.  We are all unique.  The sub clade of L that this page focuses on is L1b.  All male carriers of L1b will carry a SNP (Single Nucleotide Polymorphism) on their Y-DNA that has been designated as M317. This SNP will be downstream of another SNP that has been designated as M22.  Finally, a Y-DNA can be said to have a terminal SNP.  A terminal, refers not to the Haplogroup (in this case L), but can be used to define right down to the last SNP on the Y-DNA, that is shared with others on a record.  If someone for example, carries Y-DNA that is proven (or predicted by comparison) to be Y Haplogroup L, and to carry M317, then their Y terminal could be designated as L-M317, or alternatively, as L1b.  This is also sometimes referred to as a haplotype.  However, a haplotype can also refer to a particular STR.

Y Haplogroup L M20

The above image illustrates a modern day distribution of Y Haplogroup L (M20) as proposed and created by Anthropogenica user Passa.

Y Haplogroup K formed from Y Haplogroup IJK in the Y-DNA of hunter-gatherer fathers and sons, that share a MRCA (most recent common ancestor) during the Upper Palaeolithic, circa 45,400 years ago.  Where did these Y ancestors live at that time?  We think that they lived in Western or Southern Asia.  Iran is a favourite proposal. Earlier Y ancestors had most likely exited Africa 20,000 years earlier, and were well established in Asia.  They had most likely met and confronted another archaic human species, The Neanderthal. This was however, a time of great expansion by humans.  The first anatomically modern humans had recently entered Europe, while other modern humans were arriving in Australia.  The Ice Age was in a flux, but glaciation was advancing.

The most recent common Y ancestor to carry Y Haplogroup LT lived circa 42,600 years ago.  Then a mutation in the Y-DNA lead to the formation of Y Haplogroup L, with a most recent common ancestor 23,200 years ago, close to the time of the Last Glacial Maximum, when ice sheets were reaching their maximum positions.  K, LT, and early L, most likely all originated in Upper Palaeolithic hunter-gatherer populations living during the last Ice Age, in the area of modern day Syria, Iraq, Iran or Pakistan.  It was a time of increased stress on human populations, that were having to adapt to some severe environmental challenges, and may have at times faced isolation into a number of Ice Age Refuges.  Some of these Upper Palaeolithic, Ice Age hunter-gatherer refuges may have been close to the Black Sea, others close to the Caspian Sea, but they were most likely located somewhere between Eastern Anatolia, and Eastern Iran, south of the Caucasus.

L1 / L2 Divergence - the Odd L2's

The oldest divergence within Y Haplogroup L.  L1, as characterised by the SNP M22, diverged from L2, as characterised by the SNP L595.  L2 was only recently discovered, and forced an ISOGG revision of Y Haplogroup L and it's nomenclature that is still causing problems.  In this article, unless stated otherwise, I am using 2017 Nomenclature.  L2 or L-L595 is very rare, but has so far cropped up sporadically across Europe, including in Sardinia, England, and the Baltic area.

That is L2 dealt with.  However, most Y Haplogroup L falls into L1. Let us start to look at the main branches of L1.  Remember, L1 is defined by the SNP M22:

Unofficial proposed tree for L1 (L-M22) 2016.  By Gökhan Zuzigo, modified by Paul Brooker.

The Big L1 Split - L1a and L1b

As can be seen above, this split occurred around 18,400 years ago, possibly somewhere between what is now Iran and Pakistan.  The L1a branches inherit the SNP M2481, and the L1b branches inherited M317.

First of all, let's look at L1a, because although it is not my sub clade, in terms of modern day population size, it appears to greatly outnumber any other L sub clade.

Pakistan and India - Present Day Home of L1a1 and L1a2

L1a splits again into two sub clades.  The split occurred around 17,400 years ago.  L1a1 (L-M27) and L1a2 (L-M357)

L1a1 (L-M27)

Defined by SNP M27 (on older nomenclature as still used by 23andMe, this was formerly L1*) is mainly found in India, particularly South West India, and in Sri Lanka. This is perhaps the most populous modern day L sub clade, found in 15% of Indian males.  However, it is not restricted to India, and has also been found in 20% of Balochi in Pakistan, and has also been reported in Kirghiz, Pashtun, Tajik, Uzbek, and Turkmen males across Central Asia.

L1a and L1a1 (L-M27) at Birds Eye Cave, Armenia 6161 years before present.

Ancient Y DNA from the Copper Age has emerged from this location in Armenia, and included L1a, and L1a1.  This might suggest, that although very successful today in India and Pakistan, that it has a Western Asian origin.

L1a2 (L-M357)

Has defined by SNP M357 (on older nomenclature as still used by 23andMe, this was formerly L3*).  This sub clade is mainly found in Pakistan, but also Saudi Arabia, Kuwait, The Chechen Republic, Tajikistan, India, and Afghanistan.  It has been found at 15% in Burusho populations, and at 25% in Kalash populations.  It is much more common in Pakistan than in India.

So, the L1a sub clades - spreading down into Southern Asia, and accounting for potentially millions of Y Men there.  Far more than any other branches of Y Haplogroup L.  However, Southern Asia is unlikely to be the origin of L.  That origin is more likely, as stated earlier, to be the place with the most diversity in branches.  That points more towards again towards Western Asia.  It's just that ancient carriers of L, appear to have been particularly successful in Southern Asia, and to have fathered more sons there.

L-M317 or L1b of Western Asia

We now move onto the branches of particular interest to myself, because I carry a Y Haplotype that belongs here.  L1b is defined by the SNP M317, that formed circa 18,400 years ago, most likely in the area of modern day Iran, or elsewhere in Western Asia.

Phylogenetic tree of L1b by Anthrogenica user Caspian (with permission):

Click on above hyperlink for full sized image

L1b is mainly distributed across Western Asia, from modern day Turkey, across to Pakistan.  However, as we will see, it also spreads in low densities across parts of Europe.  it is very much, the "Western L".

The Next split - L1b1 or L-M349.  The Levant, and Europe!

Around 14,000 years ago, another split occurred in the L1b (M317) branch. A new SNP, M349, defined L1b1.  Today, L1b1, or L-M349, is found in Western Asia, in Lebanon, Syria, Turkey, Armenia, etc.  However, it is also found scattered in low densities through parts of Europe.  It crops up in South Europe, often close the the Mediterranean Sea, including particularly in parts of Italy.  It also forms a light cluster in Central Europe.

A working map of Y haplogroup L sub clades by Edward Chernoff.  This map is incomplete, but is published here with permission of Edward Chernoff.  Copyrights applied.

Branching away from a common Y ancestor with L1b1 (M349), is another 14,000 year old line defined by SNP SK1412, L1b2.

L1b2 (L-SK1412) splits - Pontic Greeks, and the others...

13,000 years ago,, during a cold stage towards the end of the last Ice Age, the L1b2 (SK1412) Y branch divides again.  Very recent research suggests that it split into three lines: L-SK1415 (L1b2a), L-PH8 (L1b2b), and L-SK1414 (L1b2c).

L1b2a (L-SK1415), has as far as I know, only been detected in a Makrani Balochi survey in SW Pakistan.

L1b2b (PH8), is found in Turkey, Greece, Armenia, Chechen Republic, Iraq, etc.  It is associated particularly with the Pontic Greek ethnicity from Eastern Anatolia, and around the Black Sea.  A further division within PH8 has been detected at around 3,000 years ago.

Finally ... mine:

L1b2c (L-SK1414, FGC51074), has so far been SNP detected only in Makrani Balochi, in SW Pakistan, Lebanon (Druze), and in England. STR predictions for L-SK1414 have also been found in France, Italy, Saudi Arabia, Iran, Turkey, and Portugal.  In addition to SK1414, I have with the assistance of Gareth Henson, a FT-DNA Big Y test, accompanied by further analysis of their raw data, by Yfull, and FullGenomes, ascertained 117 novel SNPs looking for first time matches.  As can be imagined, I'm very keen that further L Y-Men should test.

Those tentative European Y haplogroup L links

We have seen above, that again, and again, Y haplogroup L (M20), and several of it's sub clades appear to have Western Asian origins, despite success of some of those sub clades today in India and Pakistan.  Y haplogroup L has not been linked to the Yamna hypothesis, that has taken credit for the origin of many haplogroups that are successful today in Europe.  Y-DNA L was located to the southern side of the Caucasus, between present day Turkey and Pakistan.  However, two particular Y-DNA L sub clades do make mysterious appearances across Europe.

1) L-L595 (L2) has only recently been discovered, so far, exclusively across Europe, in very low numbers.

2) L-M349 (L1b1), downstream of M317, also spreads across South Europe, and clusters at the Rhine-Danube.  I have on 23andMe forums, seen a number of testers that unfortunately have not tested their Y elsewhere, claim Ashkenazi paternal ancestry, but this is far from common to all European L-M349 samples. Although rarely forming much more than 1% of all Y along the Mediterranean coast of Southern Europe, this percentage does occasionally rise higher, for example, in parts of Italy.

When did L2 or even L1b1 enter Europe?  L2 has only so far been found in Europe.  There are some suggestions that some European L could be survivors from the Eurasian Neolithic.  However, ancient DNA has not yet been found to support this hypothesis. 

Prime resources

L Yfull Tree

https://www.yfull.com/tree/L/

Wikimedia entry for Y Haplogroup L-M20

https://en.wikipedia.org/wiki/Haplogroup_L-M20

FTDNA L The Y Haplogroup L Project

https://www.familytreedna.com/public/Y-Haplogroup-L/

Marco Cagetti's Y Haplogroup L

http://www.cagetti.com/Genetics/L-haplogroup.html

Anthropogenica Y Haplogroup L Forum Board

http://www.anthrogenica.com/forumdisplay.php?37-L

ISOGG 2009 Y Haplogroup L (Useful for understanding 23andMe Y haplogroup result of L2*)

http://isogg.org/tree/2009/ISOGG_HapgrpL09.html

ISOGG 2017 Y Haplogroup L

http://isogg.org/tree/ISOGG_HapgrpL.html

Other resources

Europedia Y-DNA Haplogroup L

http://www.eupedia.com/europe/origins_haplogroups_europe.shtml#L

23andMe users should note that the company in 2016, still used a very outdated ISOGG nomenclature system.  My 23andMe reported haplotype was L2*.  However, using ISOGG 2016, this is now L1b (L-M317).  NOT to be confused with modern day L2 (L-L595).

Facebook Y Haplogroup L Group

https://www.facebook.com/groups/773887796013634/

L-M317 STR Alpine cluster article

https://figshare.com/articles/L_M317_STR_marker_likelihood_tree_focuing_Alpine_cluster/105684

Familypedia Wiki for Haplogroup L

http://familypedia.wikia.com/wiki/Haplogroup_L_(Y-DNA)


The 23andMe DNA results are in!

The results were uploaded to my 23andMe profile today.  I posted/registered the sample from the UK, nine weeks ago.  The sample traveled to the USA lab via a NL holding depot.  It took six weeks to process the sample and results, from the time of being marked as arriving at the USA lab.  I feel very fortunate, as many 23andMe customers are reporting a seasonal log-jam that is delaying the process.  My results though were comfortably within the proposed time frame.

There were a number of pleasant surprises.  The results were far from boring.  

Genetic Risk Factors

On the health side that we UK customers can presently still enjoy - there was only good news.  Although I have a family history of Alzheimer's that is strong on my father's side, there was no identification of any genes in my DNA, that have so far been associated with increased risk of the illness.  If my father did have these genes, I didn't receive them.  It does not mean that I will never be at risk to the illness, but it gives me some comfort.  Indeed, all of my 23andMe genetic risk factors were good.  There was no bad news.

Traits

An amusing little trait, that IS identified by the DNA analysis, is on Asparagus Metabolite Detection.  When I eat asparagus, my urine smells strongly.  It confirms for me - that the system works!  It also correctly identifies that I have a sweet tooth, that I have blue eyes, etc.

Now to the genetic genealogy goodies.

Ancestry

Y-DNA

The genetic marker that I inherit from my strictly paternal lineage - father's father, father, and so on, going back.  On paper, I've traced this back to a John Brooker, that lived in Oxfordshire, but was born outside of that county, perhaps in nearby Berkshire, circa 1785.  Of course, that is if no-one ever lied in forms over who the father was.

This one was a shocker.  A little background first.  Although my paper ancestry over the past 350 years is overwhelmingly localised in parts of the county of Norfolk, in East Anglia, my paternal-line surname carrier, that should be the donor of my Y chromosome marker, or Y-DNA, can be traced to Oxfordshire, in Wessex.  Out of my eight paper great grandparents, seven were Norfolk born and bred.  However, the exception was my paternal great grandfather.  Therefore I would not expect my Y-DNA to belong to any local Norfolk gene-pool.  It is the least representative lineage for my heritage.  This is why I feel that people can sometimes place too much value on their haplogroups.  I did however, expect it to belong to a common English or British haplogroup such as the Y-DNA R1b group.

I was in for a surprise.  It is exotic L2*.

From initial research including an Internet search, this haplogroup forms only a rare back scatter across Europe.  It appears more commonly across Western Asia and the Sub-Continent, from Turkey to Southern India.  It is most common in Pakistan, where it may originate, circa 30,000 years ago.  It is not a common European Y-DNA haplogroup.  I need to more carefully research this in the near future, but I'm in awe to find that I have an exotic Y-DNA.  It does conjure up images of one of my paternal ancestors being a Syrian archer, or Persian mercenary in the Roman Army, fathering a child, while stationed in Britannia, or perhaps elsewhere in Roman Europe.  But that might be too fanciful.  Anyway, I'm having pheasant curry for dinner tonight.

This genetic marker should be shared with my son, and my brother.  A few of my first cousins will also have it.

mt-DNA

The genetic marker that I inherit from my strictly maternal lineage - mother's mother, mother and so on back.  On paper, I've traced it back to a Mary Page, who was born in 1802, in Norfolk.  I like the maternal line, as it is actually the most biologically secure.  Few forms lie about who the mother is.  I'd expect my mt-DNA to be a haplogroup firmly established in East Anglia.

A nice one to have.  It is H6a1.

This haplogroup belongs to the Helena group.  However, it is not ancient European.  H6 is believed to have mutated from H around 30,000 years ago in Central Asia.

H6a1 has recently been associated with the Yamnaya migration into Western Europe, from the Eurasian Steppes to the north of the Black Sea, some 4,000 to 5,500 years ago.  In Europe itself, it could be associated with a number of Early Bronze Age cultures, the Corded Ware culture.  It has been linked with the R1b Y haplogroup, that dominates Western European countries such as Ireland, France, and the British Isles.  Recent studies have indeed suggested a significant displacement of people in Western Europe, that occurred in late prehistory, with the arrival of pastoralists from Eurasia.  This migration is also associated with the rise of the dominant Indo-European linguistic group of Europe.  If H6a1 does indeed prove to be linked to the Indo-European explosion of the early Bronze Age, I'd be very happy.  I like to imagine one of my maternal ancestors 5,500 years ago, accompanying a band of prehistoric pastoralists, that are heading westwards into Europe with their horses.

This genetic marker will be shared with my mother, my brother, my sisters, and their children.  A few cousins will also share it.

Ancestral Composition

This is an area that I've been trying to understand recently.  It uses computer analysis, to compare my autosome DNA to a number of others in reference populations from around the World, which then composes suggested ancestry in percentages.  This magic attempts to look not at a few genetic markers or haplogroups, but at all of the patterns in my autosomal DNA, to predict likely ancestry on any lineages that survive in my DNA.

Previous to receiving my results, I recently revised and bolstered up my paper genealogy based family tree,  I now have 172 direct ancestors listed, going back to Generation 14 during the 17th Century.  I noted that all, and everyone of my paper recorded ancestors were English.  All of them.  That includes all of my eight grandparents, all of my sixteen great great grandparents, and thirty of my thirty two great great great grandparents.  That is 100% English.

Now, I'm sure that you'd agree, I should be expecting my 23andMe ancestry composition to give 100% English, right? Well no.  They can't presently identify an ethnic group like the English.  Instead, I should expect my results to fall 100% into the British & Irish category.

100% British & Irish?  No, I'll give this one early.  it was 32% British & Irish on speculative mode.  More on this further down.

My paper research before I received my results also revealed just how concentrated, most of my ancestry has been over the past 350 years.  I compiled the below map of East Anglia.   The BLUE marking the places of ancestral events from my family tree on my father's side; and the RED marking the places of ancestral events on my mother's side.  The larger the marker, the more events recorded.

I also made a map based on East Norfolk during the 4th Century AD, before sea levels fell, and drainage changed the coastline.  I then marked out the area of my mother's ancestry on that.

 The point that I was trying to make was that I believe that my ancestry may have been more exposed to the North Sea Immigration waves of the 4th to 11th centuries AD.  More exposed than your average person of British & Irish heritage.  I also suggested that East Anglia, very much a part of the North Sea World, was particularly attractive to Early Medieval migrants from Frisia, Schleswig-Holsten / Angeln, North Saxony, and from Denmark.

On reviewing the 23andMe DNA Ancestry Composition of an admittedly small sample of other users with strong English heritage, I concluded that the average ethnic English person receives the results:

100% European

60% British & Irish

10% French & German

2% Scandinavian

25% unidentified broadly NW European

People of Irish heritage, or even Americans with either Irish or British ancestry, tend to score a higher percentage of British & Irish than do the present day ethnic English.  23andMe has a generous and growing reference population in it's British & Irish database.  However I hypothesised that 1) the 23andMe B&I reference is skewed to the Irish, and away from English.  It is also possible that it is distorted by a case of genetic drift by testing Americans of British origin.  2) that the British & Irish designation may actually be inadvertently looking at DNA that arrived in the British Isles largely previous to the early medieval North Sea migrations.  To the British and Irish genes that have been here since late prehistory.  On the other hand, the French & German, the Scandinavian, and perhaps some of the undesignated Broadly NW European percentages that are usually assigned to the ethnic English, may actually reflect early medieval migration from across the North Sea.  The computer analysis is simply unable to distinguish some of the DNA from that of present day French, Germans, or Scandinavians, because of ancient admixture.

I'm told that this would not be the case, that 23andMe ancestral composition could not detect such deep, ancient admixture.  However, what if I am correct about my own heritage - that I likely have enhanced levels of Anglo-Saxon and perhaps Norse heritage, because of the geographical location of so many of my ancestors?  Should I not expect even lower percentage of the 23andMe British & Irish category, and even higher percentages of other NW Europeans from across the North Sea?  So what was my 23andMe ancestry composition percentages (speculative mode)?

100% European.  Broken down into:

94% NW European.

3% South European.

I'll get to the South European later, but what about this North west European?  Let's break it down into 23andMe's sub categories:

32% British & Irish

27% French & German

7% Scandinavian

29% undistinguished broadly NW European

Oh my goodness.  It correctly fits my prediction.  I have more than double the average percentage of F&G and Scand for English people.  Despite having a paper researched genealogy that is 100% English, 23andMe's ancestry composition based on a generous reference sample size of 1251 sets, gives me 32% British & Irish.

So a predicted, but still incredibly exciting result.  I'm chuffed to bits.  It does in my eyes, blow 23andMe's British & Irish designation out of the water though.  Their reference samples do not appear to match the East English.  Instead, their software misreads some of the English DNA for French & German, or Scandinavian.  I'm suggesting that this is because of ancient admixture, during the 4th to 11th centuries AD, with North Sea immigration.  I invite others to knock my suggestion down.

One more surprise from my Ancestry Composition:  A South European 2.7%.  Broken down into 23andMe's sub categories:

0.5% Iberian

2.4% undistinguished broadly South European

This looks real.  It appears that I have a small percentage of South European heritage.  Most likely from Spain, Portugal, or Basque.  I probably have Iberian ancestry that I have not yet detected using paper genealogy.  Either that, or it's an anomaly, a incorrect interpretation.

Neanderthal Ancestry

Finally, how much Neanderthal DNA do I have?  How much of my DNA was shared by the archaic humans that lived across parts of Eurasia, between 350,000 to 30,000 years ago?  Evidence of early admixture events between Neanderthal and anatomically modern human populations?

An estimated 2.9%.

That's just slightly above the average of 2.7% for modern Europeans.  So I am not more Neanderthal than most others.  Sorry to disappoint.

All in all, very happy that I spent the money.