G-L91

This is the home page for men who have the L91 mutation or are predicted to have it.
It is also temporary home for those L91 subgroups defined by sharing a marker value
oddity.  These will have their own pages when whole sequencing identifies permanent
SNP mutations to define them. We also do not know how these marker-defined
subgroups relate to each other.

L91 has two subgroups defined by SNP mutations FGC2315 and PF3239.
The three marker defined subgroups have DY395s1=15,17, DYS425=null & 
      DYS607 greater than 15.

M201
      P287
            P15
                 L1259
                        PF3146
                                PF3177
                                      M286
                                       L91

The diagram just above shows where L91 is located in the G Tree.  It is a
branch of PF3177 and younger than PF3177.   There is a page titled G-PF3177.
PF3177 is younger branch of PF3146 which is the mother group of PF3177.  Much 
more information is known of this overall PF3146 group and explained at G-PF3146 
on the left.  PF3146 is more recent than L1259 & P15 which are the mother groups of P3146.  
They cover more men. There is a separate page for P15, shown in the left tabs here 
as G-P15. But the page for the even older SNP, G-P287, covers yet more persons, 
and its page has extensive information about that broad group.  And finally these 
all belong also to P287's mother group, G-M201, which has even more information 
for that M201 mutation which is found in every single G man.

M286 is a sister group of L91, both branching independently from PF3177.

There are several marker features seen in L91 men of all types and subgroups. 
(1) They are more likely than the average G man to have a value higher than 29 for
the second component of DYS389.  A double value for the DYS19 marker can occur
throughout the overall PF3146 group, not just in L91, but is rare outside the
PF3146 group. When one of the double 19 values mutations to the same value as 
the other, the lab can no longer identify the double value.

L91 was first identified at Family Tree DNA in April 2009.

Age of L91

Full sequencing of L91 men will give the most reliable information about the
age of L91 and its subgroups.  Until then, we must use 67-marker comparisons
to provide vague estimates.  We do have dating from Oetzi, the Iceman mummy,
who belongs within the L91 branch.  And that dating shows that L91 is at least
the age of the Iceman, about 5000 yrs old.   In reality is likely much, much older
There is extensive variation among L91 samples for 67 markers, with men from
other haplogroups showing up at the outer end of the timeframe.

The DYF395s1=15,17 subgroup can be much more reliably dated to a common 
male ancestor who lived within the last 500 years.  The DYS425=null subgroup
is considerably older, probably in excess of 4000 yrs.  And the DYS607 greater
than 15 men probably share their common ancestor at 2000 to 3000 yrs ago.  In 
each case, this is the most recent common ancestor.  There may have been 
other, older ancestors who shared the same mutations, extending the age of
the subgroup.

Geographic Distribution of L91

The L91 men so far negative for L91 subgroups are so far found in men with
n.w. African or western European ancestry.

The DYF395s1=15,17 subgroup is so far found only among men from Ireland
and England.

The DYS425=null subgroup is still fragmentary, and its members report 
ancestry in Ukraine, England and Hungary.

The DYS607 > 15 subgroup members report ancestry in Germany or the
Netherlands so far.

Keller et al. in 2012 showed that L91 has an unusual concentration on the
islands of Corsica and Sardinia off the western coast of Italy.  The later
publication by Francalacci et al in 2013 corroborated this and provided
SNP info from low coverage sequencing.  Preliminary info is that 
2/3 of them comprise a series of new subgroups which may not have
much prevalence outside of Sardinia and Corsica.

Berger et al. in 2013 in a large sampling of the Austrian population found
about 4% of all G persons there were L91+.  L91 represented just under
a half of one percent of the total Austrian population and showed no
special concentration in the area just north of where the L91+ Oetzi mummy
was discovered.

Origins of L91

Because L91 is so old, it is not clear at all how migrations took place.  It is
even possible that all the major migrations took place during Stone Age
times when hunter-gatherers roamed about without anyone recording
their movements.  Whole sequencing will allow better time estimates of
when branches split and this indirectly will point to when and where 
migrations occurred.

Even with the three subgroups listed here, we only know where men live
today -- not how their ancestors arrived there.

Maps of L91 Men

A map of L91  project men is found at:
Look at top left of map and choose the L91 selections which are abt 40 or more
items down the list.

The map from the Keller et al. study in 2012 is at
(figure 3 there).  This map was created using 7,797 European samples from
all haplogroups.  There was no information as to how many were from 
Sardinia and Corsica which show especial L91 concentrations.

Source Information

Keller and others in 2012 were the only authors ever to test for just L91

Rootsi and others provided L91 markers samples and calculated a
coalescent time for them.

Berger and others in 2013 found small amounts of L91 in a large sampling of the 
Austrian population.

Francalacci and others in 2013 showed the high frequency of L91 men in Sardinia
supplement has info


Rootsi et al.  provided these L91 samples in 2012:  (marker names abbreviated, some markers, some samples omitted)
           These are not tested for some of the L91 subgroups known to us.

Druze of Levant      14       22   15,16 10    13-13    12    14-32   11   17   22
Iranian Azeri           14      21   14,15 10    12-13     12   12-28   11   17   22
Iranian Azeri           14      22   14,17 10    13-16     12    14-31  11   16   22
Iranian Azeri           13      24     15   10     13-14     12   13-30   11   19  21

Special Testing

The following L91 men negative for all L91 subgroups or untested for all 
subgroups have the following results for newer, poorly defined SNP mutations:   
                                      Red items are L91's SNP subgroup + means positive, - means negative

L31+ Aboulhamid, Maranas
L78- Aboulhamid, Maranas
L89- Maranas
L90- Aboulhamid
L154+ Maranas
L156+ Aboulhamid, Maranas
L166- Aboulhamid, Cline, Heubach, Layton, Perez, Whitehouse
L167- Aboulhamid, Cline, Heubach, Layton, Perez, Whitehouse
L190- Aboulhamid
L224+ Maranas 23&Me
L225+ Maranas 23&Me
L382+ Maranas 23&Me

The following DYS607 greater than 15 men have the following results for newer, 
poorly defined SNP mutations:   
                                      Red items are L91's SNP subgroup + means positive, - means negative
L166- Roukens
L167- Roukens
L293- Cline

The following DYS425=null men have the following results for newer, 
poorly defined SNP mutations:   
                                      Red items are L91's SNP subgroup + means positive, - means negative
L166- Giddens, Toulouse
L167- Toulouse

         more info about these SNPs can be found at:
                   type L1325, for example, in the Landmark box.  After searching, look
                               for the small L1325 in the colored area and click on it.
Comments