On Thursday, January 12, 2017 at 1:30:49 PM UTC-6, Richard Smith wrote:
> On 12/01/17 18:56, taf wrote:
> > On Thursday, January 12, 2017 at 10:23:35 AM UTC-8, wjhonson wrote:
> >> If every living person were atDNA tested, would we be able to reconstruct the DNA of Charlemagne?
> >
> > No.
> As Todd says, not remotely. As far as I'm aware, only two of
> Charlemagne's children have credible descents to modern times: they are
> Pepin of Italy, and Louis the Pious. Each would have inherited roughly
> half of his DNA from Charlemagne. This means that we expect roughly a
> quarter of Charlemagne's DNA not to have been inherited by either of
> these two children. That part of his DNA is essentially unknowable.
>
> A further quarter of his DNA was only inherited by Pepin of Italy. The
> only documented descent to modern times from Pepin is through his son
> Bernard of Italy. Bernard only inherited half of his DNA from Pepin, so
> that's half of the quarter of Charlemagne's DNA that only Pepin
> inherited lost. That's now three-eights of his DNA lost.
>
> We can keep doing this, and at each step more DNA is lost.
>
> There are no documented descents from an ancestor of Charlemagne that
> does not pass through Charlemagne himself, so there's no possibility of
> inferring this lost DNA from them.
>
> At best, possibly a few fragments of his DNA have survived to present
> and could be identified through detailed autosomal DNA testing of the
> whole population.
>
> Richard
I do not know a ton about genetics, but I believe because of the random way that dna or segments of dna can be present in one person (sibling,cousin etc.) and not another person even when they are closely related, and even though we only inherit 50% from each parent I think if enough of his descendants were tested it might be possible to piece most of it together. They have come so far with dna that they were able to estimate my y-dna haplogroup even though I havent tested.

On Saturday, March 11, 2023 at 7:55:41 AM UTC-8, T. Snell wrote:
> On Thursday, January 12, 2017 at 1:30:49 PM UTC-6, Richard Smith wrote:
> > On 12/01/17 18:56, taf wrote:
> > > On Thursday, January 12, 2017 at 10:23:35 AM UTC-8, wjhonson wrote:
> > >> If every living person were atDNA tested, would we be able to reconstruct the DNA of Charlemagne?
> > >
> > > No.
> > As Todd says, not remotely. As far as I'm aware, only two of
> > Charlemagne's children have credible descents to modern times: they are
> > Pepin of Italy, and Louis the Pious. Each would have inherited roughly
> > half of his DNA from Charlemagne. This means that we expect roughly a
> > quarter of Charlemagne's DNA not to have been inherited by either of
> > these two children. That part of his DNA is essentially unknowable.
> >
> > A further quarter of his DNA was only inherited by Pepin of Italy. The
> > only documented descent to modern times from Pepin is through his son
> > Bernard of Italy. Bernard only inherited half of his DNA from Pepin, so
> > that's half of the quarter of Charlemagne's DNA that only Pepin
> > inherited lost. That's now three-eights of his DNA lost.
> >
> > We can keep doing this, and at each step more DNA is lost.
> >
> > There are no documented descents from an ancestor of Charlemagne that
> > does not pass through Charlemagne himself, so there's no possibility of
> > inferring this lost DNA from them.
> >
> > At best, possibly a few fragments of his DNA have survived to present
> > and could be identified through detailed autosomal DNA testing of the
> > whole population.
> >
> > Richard
> I do not know a ton about genetics, but I believe because of the random way that dna or segments of dna can be present in one person (sibling,cousin etc.) and not another person even when they are closely related, and even though we only inherit 50% from each parent I think if enough of his descendants were tested it might be possible to piece most of it together. They have come so far with dna that they were able to estimate my y-dna haplogroup even though I havent tested.

It's good out about 3rd or 4th cousin (4 or 5 generations of ancestors).

On Sat, 11 Mar 2023 07:55:39 -0800 (PST), "T. Snell"
<tesg0227@gmail.com> wrote in soc.genealogy.medieval:

>I do not know a ton about genetics, but I believe because of the random way that dna or segments of dna can be present in one person (sibling,cousin etc.) and not another person even when they are closely related, and even though we only inherit 50% from each parent I think if enough of his descendants were tested it might be
possible to piece most of it together. They have come so far with dna that they were able to estimate my y-dna haplogroup even though I havent tested.

A standard DNA test will sample about 600,000 values out of
3 billions. It seems there are about 1 million values in all
could be used, perhaps 2 to 3 millions as a genetician told
me recently. So you have at most 3 millions values for someone
living today to get a reference.

The average is 50% from each parent, actually more 40 to 60
if you look at 2 generations. But lets take 50%.

Suppose 40 years per generation, which is very optimistic.

The inherited DNA values would be like in this table:

3000000 2023
1500000 1983
750000 1943
375000 1903
187500 1863
93750 1823
46875 1783
23437,5 1743
11718,7 1703
5859,38 1663
2929,69 1623
1464,84 1583
732,42 1543
366,21 1503
183,11 1463
91,55 1423
45,78 1383
22,89 1343
11,44 1303
5,72 1263
2,86 1223
1,43 1183
0,72 1143

Suppose we use at least 200 contiguous values. In the best
case, we may identify a pattern formed in the 1400s, far from
year 800. With more realistic values, 30 years/gen, this is
200 values is reaches in the 1600s.

Y DNA is very different because there is no combination for
each generation. The average is 1 mutation per 82 years. So if
we have 2 male descendants of Charlemagne, then the average
would be 15 mutations out of 1 million of good values found in
the Y chromosome. Not the same thing at all.

Denis

P.S. The Y DNA haplogroup is presumed from other testees. No testee
then no haplogroup. And the universal average in christian tradition
would be 1% NPE, i.e. 1/100 of legal fathers are not the bio father.

Record is close to year 1000 from some noble lineages. So it is
technically possible to get the Y DNA of Charlemagne, but you will
need 2 complete lineages to him from living persons, and thus far,
none is known.

--
Denis Beauregard - généalogiste émérite (FQSG)
Les Français d'Amérique du Nord - http://www.francogene.com/gfan/gfan/998/
French in North America before 1722 - http://www.francogene.com/gfna/gfna/998/
Sur cédérom/DVD/USB à 1790 - On CD-ROM/DVD/USB to 1790

On Saturday, March 11, 2023 at 7:55:41 AM UTC-8, T. Snell wrote:
> On Thursday, January 12, 2017 at 1:30:49 PM UTC-6, Richard Smith wrote:
> > On 12/01/17 18:56, taf wrote:
> > > On Thursday, January 12, 2017 at 10:23:35 AM UTC-8, wjhonson wrote:
> > >> If every living person were atDNA tested, would we be able to reconstruct the DNA of Charlemagne?
> > >
> > > No.

> I do not know a ton about genetics, but I believe because of the random way that
> dna or segments of dna can be present in one person (sibling,cousin etc.) and not
> another person even when they are closely related, and even though we only inherit
> 50% from each parent I think if enough of his descendants were tested it might be
> possible to piece most of it together. They have come so far with dna that they were
> able to estimate my y-dna haplogroup even though I havent tested.

I previously gave short answers, but here is a really long one.

'Come so far' has nothing to do with it. The problem cannot be solved by scientific progress because the problems are not related to scientific capabilities, they are statistical and demographic.

The dual problems are, first, the survival of the DNA to modern times, and second, identifying that DNA. As Denis already has described today, in a single line the amount of DNA deceases by 50% per generation, and so the chances of any descendant of Charlemagne actually having Charlemagne's DNA is slim. However, there are probably hundreds of millions of descendants of Charlemagne, so one would think that most of the DNA should survive somewhere among this group. However, most of these are unusable - even for a much closer relationship, you need corresponding genealogy, and most of the descendants of Charlemagne do not know and cannot be traced to him.

Even if they all had pedigrees, there is a problem that would prevent complete reconstruction of Charlemagne's DNA. Unless I am forgettign something, Charlemagne's only known descent comes via two children, Pepin and Louis. Each would have 50% of Charlemagne's DNA, and would be predicted to share half of that - 25% of Charlemagne's DNA would be carried by both; 25% would be unique to each, adn 25% would be carried by neither - this would be lost, impossible to ever determine. Further, one of those lines, the Italian line, has several successive generations with only one known child with descent. In each of those generations, that 25% unique to the son, Pepin in this case, would be again divided by half. While there is speculation of others, all proven descnts from Pepin (unless there is one not coming to mind) come through Heribert I of Vermandois, so that means the 25% unique to Pepin is reduced to just over 3%. Louis doesn't have the same problem, having descent through multiple children but there is still going to be some loss, so taken together, we have already lost half of Charlemagne's DNA before things start seriously branching, and it can never be recovered, and that's before we even start doing any actual science. With the progressive losses of DNA over subsequent generations, the proportion of Charlemagne's DNA still around in anyone is small and many people with Charlemagne's DNA will have the exact same piece, a piece so small that it will also be shared by a much larger cluster of descendants of some 5000 year old neolythic farmer or hunter-gathrer who had it before Charlemagne, while most Charlemagne's DNA will have been lost entirely (or may be carried by Charlemagne's descendants but not through Charmemagne's line, instead coming a different way from that hunter-gatherer, leading us . . . ).

Now let's look at the second problem. How you go about identifying DNA from a specific ancestor over the short range is to identify and test all of the known (via standard genealogy) descendants of the target ancestor. (There is a more tedious approach one can do next, progressively reconstructing the genomes of the intervening generations until you get back to the target and their spouse, without being able to tell which is which.) You then test known descendants of ancestors of the target and matches between target descendants and target-ancestor descendants, to distinguish which target-descendant DNA would have come via the target. That is how you identify just a small part of the genome of the target. As pj already mentioned, we have no known descents from Charlemagne's ancestors. Even if you accept some of the assumptions made about kinships in his tree, you would need a descendant of an ancestor of Charlemagne that you know doesn't also descend from Charlemagne, or you couldn't distinguish whether the shared DNA came Charlemagne's lineage and not from his wife. We need someone we KNOW does not descend from Schalemagne, yet who is descended from one of his ancestors, and we can't even be sure that any given member of the European diaspora is NOT descended from Charlemagne - indeed, statistical models predict they all are. The very pervasiveness that enables a lot of descendants to be tested deprives us of any utility of the results.

And that brings us to the other big problem. The whole process is predicated on the assumption that the people being tested are only connected via the descent of interest. Modelled on a modern diverse highly-mobile population, this is a viable assumption, but it is an assumption. However, if they are related in any other way, then one could get a coincidental match that has nothing to do with the target line, and hence falsely identify shared DNA as belonging to the target. In the more insular and inbred communities of the 18th century, this assumption loses all validity. Over a 1200 year time span, anyone descended from Charlemagne is also going to be descended from essentially every other identified person of the time in Europe with known descendants. The shared DNA could come from any one of them. This also stands in the way of a more tedious generation-by-generation reconstruction, which not only requires no other shared ancestors, but they would also need only to have a single descent from the common ancestor, Charlemagne, but everyone descended from Charlemagne once is descended from him many times over, thereby poisoning generation-by-generation reconstruction. That leaves descent from shared ancestry as the only means to identify relevant DNA, and as already discussed, we don't have that. At this depth, we are talking about tiny pieces shared of DNA that could come from a common ancestor 1200 years ago or 12,000 years ago, with no ability to say Charlemagne is the one it came through.

As Denis pointed out, the uniparental DNA markers are easier to trace deep, but for that it absolutely requires male-line descents from multiple sons of a male target, or female-line descents from multiple daughters. We don't have that here, so it is a dead issue.

It all just doesn't work, and no amount of scientific progress will resolve these issues, that have nothing to do with the science. The only way to identify Charlemagne's DNA would be to locate a bone from him and recover ancient DNA from it. Getting it through genetic genealogy is a non-starter.

taf

On Sat, 11 Mar 2023 11:16:07 -0800 (PST), taf <taf.medieval@gmail.com>
wrote in soc.genealogy.medieval:

>The dual problems are, first, the survival of the DNA to modern times, and second, identifying that DNA. As Denis already has described today, in a single line the amount of DNA deceases by 50% per generation, and so the chances of any descendant of Charlemagne actually having Charlemagne's DNA is slim. However, there are probably
hundreds of millions of descendants of Charlemagne, so one would think that most of the DNA should survive somewhere among this group. However, most of these are unusable - even for a much closer relationship, you need corresponding genealogy, and most of the descendants of Charlemagne do not know and cannot be traced to him.

Gedmatch has a tool called Lazarus (in the Pier 1 tools).

According to this tool, you can retrieve up to 80% of DNA
of a late parent, if you test 3 children, i.e. 80% is the
optimal result in the best conditions.

So beginning from someone living today, even by remapping the
DNA available, you arrive quickly to a point where you can't
find anything. One major issue of the Lazarus tool is to know
from which chromosome is coming some value. It seems the 80%
limitation is from that issue.

Now, suppose you have some unlimited funding and you can do some
WGS (Whole Genetic Sequencing) for anyone you may need, if we
consider one limitation of the sampling system (600k/3G), I
suppose the 80% can be improved, but even there, you will probably
get some limitation anyway.

>It all just doesn't work, and no amount of scientific progress will resolve these issues, that have nothing to do with the science. The only way to identify Charlemagne's DNA would be to locate a bone from him and recover ancient DNA from it. Getting it through genetic genealogy is a non-starter.

FTDNA introduced a new tool recently, called Discover. From this
tool, you can see the Y DNA found in archeologic sites. Look at
this one, R-Z381 being the haplogroup of the Bourbon:

https://discover.familytreedna.com/y-dna/R-Z381/tree

The house of Bourbon is near R-DF98 (this SNP alone will show
a much smaller set of archeological results).

But there are many known male descendants of Bourbon and no known
male descendant of Charlemagne.

Denis

--
Denis Beauregard - généalogiste émérite (FQSG)
Les Français d'Amérique du Nord - http://www.francogene.com/gfan/gfan/998/
French in North America before 1722 - http://www.francogene.com/gfna/gfna/998/
Sur cédérom/DVD/USB à 1790 - On CD-ROM/DVD/USB to 1790

On Saturday, March 11, 2023 at 12:16:15 PM UTC-8, Denis Beauregard wrote:

> FTDNA introduced a new tool recently, called Discover. From this
> tool, you can see the Y DNA found in archeologic sites. Look at
> this one, R-Z381 being the haplogroup of the Bourbon:

One does have to be very careful with archaeological samples. The work done on the Babenburgs almost certainly misidentified the corpses, and the initial work that identified the Bourbons haplotype based on two samples that supposedly came from Henry VI relics, but ended up bearing no similarity to any of the living Burbons when they finally got around to testing them. Of the reported work on several supposed Rurikid burials not a one of them matches another. Misidentification is a real issue, and can only be confirmed with DNA from descendants, and that brings us right back to the problem of no uniparental descendants to test. I know of only one archaeological identification made without descendants to test - Copernicus, whose skeleton was matched with a stray hair found in a book in a Swedish archive known to have once been owned by the man. That is such a rare circumstance that it is not generally applicable.

And the vast majority of archaeological samples are of random anonymous people a lot farther back in history than anyone who can be identified via genealogy. Useful if you want to trace one's paternal or maternal lines back through the cultural/tribal groups they arose from, but not for linking DNA to named individuals.

taf

On Saturday, March 11, 2023 at 10:33:54 PM UTC, Todd wrote:
> On Saturday, March 11, 2023 at 12:16:15 PM UTC-8, Denis Beauregard wrote:
>
> > FTDNA introduced a new tool recently, called Discover. From this
> > tool, you can see the Y DNA found in archeologic sites. Look at
> > this one, R-Z381 being the haplogroup of the Bourbon:
> One does have to be very careful with archaeological samples. The work done on the Babenburgs almost certainly misidentified the corpses, and the initial work that identified the Bourbons haplotype based on two samples that supposedly came from Henry VI relics, but ended up bearing no similarity to any of the living Burbons when they finally got around to testing them. Of the reported work on several supposed Rurikid burials not a one of them matches another. Misidentification is a real issue, and can only be confirmed with DNA from descendants, and that brings us right back to the problem of no uniparental descendants to test. I know of only one archaeological identification made without descendants to test - Copernicus, whose skeleton was matched with a stray hair found in a book in a Swedish archive known to have once been owned by the man. That is such a rare circumstance that it is not generally applicable.
>
> And the vast majority of archaeological samples are of random anonymous people a lot farther back in history than anyone who can be identified via genealogy. Useful if you want to trace one's paternal or maternal lines back through the cultural/tribal groups they arose from, but not for linking DNA to named individuals.
>
> taf

I often see experts on TV progs giving very definite conclusions after DNA analysis. I remember
when they did a prog on a skeleton found inside Cheddar caves dating 000s BC, and after
comparing the DNA taken from this chap with that of modern locals pronounced that they
he was related to many living in the area today. Surely theres adanger of contamination when
bones have been laying about so long and could have been handled by many people
before and after discovery? Another case I remember is when some Ancient UFO enthusiasts
got some bizarre skulls they had acquired from Peru tested, and found that the closest match
were people in Scotland! They were hoping they were alien progeny, not human at all!

Mike

On Tuesday, March 14, 2023 at 4:32:18 PM UTC-7, mike davis wrote:

> I often see experts on TV progs giving very definite conclusions after DNA analysis. I remember
> when they did a prog on a skeleton found inside Cheddar caves dating 000s BC, and after
> comparing the DNA taken from this chap with that of modern locals pronounced that they
> he was related to many living in the area today. Surely theres adanger of contamination when
> bones have been laying about so long and could have been handled by many people
> before and after discovery?

Yes, absolutely. There are techniques now available that can identify such contamination and eliminate it (or rather, to identify authentic ancient DNA and pull it out of the contaminated sample for testing. This was not available at the time the Cheddar Man analysis was done. However, contamination is not nearly as big a problem with mtDNA, which is what they studied. It is present in cells (and hence in bone, etc) at 1000x the amount of other DNA, so the chances that there will be contaminating DNA in amounts that would swamp this true signal is less (though the approaches used at the time unwittingly favored such contamination - probably not enough to make a difference in this case).

There are two dirty little secrets about this analysis. The first was a secret at the time. They actually best-matched a local child, but they didn't want to put the kid through the media wringer, so they took the next closest local match and made all of the press about him. The second is that the analysis they did would now be deemed superficial, at a level of precision such that they were identifying a mtDNA group rather than a specific haplotype, and one that we now know is found all over England. They could have tested a few dozen people anywhere from Yorkshire to Essex and they probably would also have found a 'match' at this low level of resolution. That doesn't mean descent from Cheddar Man - it simply marks descent from the earliest agriculturalists to settle England.

> Another case I remember is when some Ancient UFO enthusiasts
> got some bizarre skulls they had acquired from Peru tested, and found that the closest match
> were people in Scotland! They were hoping they were alien progeny, not human at all!

Don't know about Scotland, but a more recent analysis of one of these 'alien baby' skeletons concluded by DNA and 14C dating that it was a (probably miscarried) 19th-century child with a birth defect, belonging to the population of the area. That only proves that the conspiracy to suppress UFO evidence is bigger than we thought.

taf