From Newsgroup: soc.genealogy.medieval
On 10/1/2024 3:52 PM, miked wrote:
On Sat, 21 Sep 2024 15:27:38 +0000, Andrew Lancaster wrote:
In some studies with DNA from remains before and after the arrival of a
new group of people, the researchers look at the mtDNA and yDNA
frequencies before and after the arrival. If the subsequent yDNA is
predominantly that of the arrivals but the mtDNA is predominantly that
of the original inhabitants, the result is considered consistent
with or
indicative of a conquest by a predominantly male group of migrants.
Yes Jan that is a good point, but I feel pretty strongly these days that
Y DNA and mitochondrial can't normally be used to come to such
conclusions.
Mitochondrial DNA in particular is generally useless for predicting
where someone is from in any accurate way, either modern or ancient.
Although patterns exist, exceptions are all too common, and this type of
DNA does not change quickly enough for variants to be associated with
any types of human societal changes we can relate to. There will always
be exceptions. It was useful in the case of Richard III because an
unusual haplotype was involved, so it helped improve the odds.
i remember watching the prog about the discovery of richards body. They showed the skeleton of a man whose had scoliosis of the spine so severe
that his backbone resembled the letter S. So the shakespear tale of the hunchback king was true! Its rather extraordinary that a man so deformed could have ridden a horse in full armour leading a cavalry charge down Bosworth hill that nearly succeeded in killing Henry Tudor, but now we
have the paralympics so anythings possible. Plus the body had a lot of wounds so clearly he died violently and the way it was presented it
seems impossible to argue with the science. AIUI the body shared the
mtdna with a man in australia [?] who was descended from richards sister
but OTH I thought the haplagroup wasnt shared with the other
descendants.
The Y DNA of Richard is still an outlier. Unfortunately, they used a
test that only gave the most basic type of result - basically only
useful to distinguish which iron-age lineage one's male ancestors
derived from, so not genealogically informative except in the broadest
sense (i.e. excluding lines descended from different iron-age lineages,
but not revealing specific relationships). They had hoped to do a full
genome, which would have given much more precise results, but last I
heard they failed to get funding for it (and with the amount of time
that has passed, that is certainly the case or we would have seen the
results by now).
Y DNA is sometimes useful in cases where a whole tree of mutations can
be localized. This is unusual, but a reasonably convincing example is
the association of R1b with Bronze Age steppe dwellers entering Europe
from the east. I think there is therefore a lot more promise once it
becomes economically feasible to sequence whole genomes for novel
mutations rather than just sampling known ones. Some genealogists are
already doing this on a small scale for things like surname and clan
studies with interesting results. (Although I have not been following it
closely for a few years.) The scientific papers basically never do this
though.
I dont pretend to understand the science of these studies, but
frequently someone posts on this group [or used to] or other forums i
read, saying something like 'i know that i'm related to king thingy,
duke wotsit or gateway ancestor 1234 in 1634 becos i share the same dna sequence/hapla thing, plus i have some from clan Mactavish, and the Castafiore of Montalbano' and so on. I spose the science is evolving all
the time, but isnt the interpretation of the results rather subjective?
The degree to which the results are subjective is a matter of
perspective. A DNA haplotype is not subjective at all - the DNA has a particular sequence, and each difference allows classification into a
specific subgroup of the larger whole. However, the use of this by genealogists is often over-interpreted in and entirely subjective
manner. Take the gateway ancestor bit. If you have a particular
haplotype, then you can exclude gateway ancestors you do not descend
from, but you can never prove you descend from a specific gateway and
not, say, a brother or uncle of that man. The only exception to this is
if the family has been characterized to such an extent that specific differences have been localized to mutations that took place at specific places in the pedigree, and if one has that difference, then it can
identify a specific line to which one belongs. However, there are only a
small number of families that have been characterized to this level, and
many families do not have a sufficient number of male-line descendants
to even allow this kind of thing.
There was an prog from about 20 yrs ago called 'Meet the Ancestors'
where they dug up some bod from ancient times, i forget whether it was
iron age or more recent, and revealed a reconstruction of his face to
the locals, who all laughed cos it looked so similar to someone they
knew, who happened to be present, who was also 1 of 5 people who shared
some dna with the old bod. This was all presented by academics and scientifically backed up,
This is popular-science nonsense, but it makes for good TV. Basically,
it fails on multiple separate levels. First, the 'science' of facial reconstruction from skulls is more of an art than a science. Only the
most basic indication of features can be deduced, and the rest is left
to the artist. If it was 20 years ago, they would have guessed on such
things as eye color, hair color, skin color, nose shape, hairline, etc.
If it bore a resemblance to someone locally, that was probably because
the artist did that intentionally (this was the approach back then -
they would give the same eye color as the predominant eye color in the
area, and the same for other features).
Separate from this, humans tend to see similarities and ignore
differences. As such, the 'similar' looking person may have shared, say
the nose shape while being different in various other respects, but
because this nose shape was the distinguishing feature of the
individual, its similarity caused people to say they looked similar in
spite of numerous other features that were distinct.
Most importantly, genetics doesn't really work this way. The appearance
of an individual is based on hundreds of different genes, all
independently assorting with each generation, plus various aspects of 'nurture', both in utero, during childhood development, and during
aging. That the person who looked similar was one of five who shared a particular genetic marker with the skeleton is entirely coincidental -
the appearance of that person had nothing whatsoever to do with the fact
that they shared that genetic marker, any more than the fact that four
other people also shared the genetic marker but apparently shared no noticeable resemblance.
That it was 'presented by academics and scientifically backed up' is the
game that some scientists play to publicize themselves and their
research - they play up this stuff for the public, often using language imprecise enough that they can maintain their integrity while giving a
false impression to the audience as to the quality of the findings, but
are much more careful about their conclusions when (if at all) they
present it in scientific publications.
but in this and other progs they never
actually show what the dna proof of a relationship actually is. As i
recall it usually looks like a lot of stripes on a chart which experts
or computer programs can see patterns.
The DNA proof of relationship is referring to sharing an identical
pattern of DNA, where 'pattern' refers to different things depending on
which analysis has been done. If your DNA has certain specific sequences
at certain locations, or certain insertions or deletions, then shared
lineage can be unambiguously determined, whether that is shared lineage
with someone in modern time, shared membership in the groups descended
from the same iron-age populations, or shared descent from the same
group of mammals that survived the asteroid that wiped out the
dinosaurs. Where it becomes more tricky for the hobbyist is
understanding how precisely the DNA result allows you to place a
relationship in time - most of the early DNA relationship claims from 20
years ago simply meant they came from the same population 10,000 years
before, and not more specific relationships.
As to the 'stripes on a chart', you seem to be referring to obsolete techniques. Modern stuff is done with high-throughput nano-scale
electronics, optics and chemistry that can only be analyzed by computers producing hyper-dense datadumps that have no real visual representation. Popular-science presentations often dig out the old visuals from earlier approaches because they are communicating through visual media and have
to give their audience something to look at, but the images they show typically have no correspondence to the techniques used in the study
they are describing.
taf
--- Synchronet 3.21d-Linux NewsLink 1.2