h a l f b a k e r yNot just a think tank. An entire army of think.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
OK, so we've sequenced the neanderthal genome.
99.7% of the nucleotide sequences of the modern human and
Neanderthal genomes are identical, compared to humans
sharing
around 98.8% of sequences with chimps.
The proportion of Neanderthal-inherited genetic material is
about 1.5 to 2.1 % &
is
found in all non-African populations. It is suggested that 20
% of Neanderthal DNA survived in modern humans.
It's a good bet we'll find some of their genome that didn't
survive
in modern humans in chimps, other apes & probably elsewhere
(we're 40-50% cabbage after all).
So with CRISPR gene editing all we have to do is find the gene
sequences we want edit some human DNA & inject it into a
hollowed out ovum.
We know modern humans could interbreed so a surrogate
mother
to bring it to term isn't going to be a problem.
There may be a small problem with miscarriage for male
neanderthal carried by a modern human but we've
immunosuppressant drug treatments that can solve that, or we
just produce some females first & use some
of
them to carry the first males.
You know, I think we can do this.
Neanderthal genetics
https://en.wikipedi...eanderthal_genetics
WikipediA [Skewed, May 31 2020]
Neanderthal genome project
https://en.wikipedi...thal_genome_project
WikipediA [Skewed, May 31 2020]
CRISPR gene editing
https://en.wikipedi...CRISPR_gene_editing
WikipediA [Skewed, May 31 2020]
CRISPR
https://en.wikipedia.org/wiki/CRISPR
WikipediA [Skewed, May 31 2020]
Modern humans may have miscarried male hybrids
https://www.science...g-babies-here-s-why
Here's why. [Skewed, May 31 2020]
Whole 3 billion letters of Neanderthal genome sequenced
https://www.genome....al-genome-sequenced
in 2010 [Skewed, Jun 03 2020]
Neanderthal mitochondrial DNA sequenced
https://www.nature....human%20neighbours.
in 2008 [Skewed, Jun 03 2020]
People aren't bananas
https://www.newscie...rts%20in%20bananas.
Letter on the misleading nature of reports of genetic differences [Loris, Jun 15 2020]
[link]
|
| |
What do you do with the baby once she is born? It would be no use raising her in a middle-class developed-world household. |
|
| |
Yet we don't yet know enough about palaeolithic culture, language, and social development norms and practices... do we? |
|
| |
Perhaps give her to an "untouched" tribe in the Amazon rainforest, or the New Guinea highlands. But then when she is grown how do we keep track of her and study her? Would we then learn more than if we just studied her foster-siblings, and save money on all the genetic engineering? |
|
| |
Or do we raise her in a zoo, with animatronic neanderthal carers, and invented synthetic palaoelithic languages and behavioural norms? |
|
| |
Reminds me of 'The Ugly Little Boy'. I don't think it'd
be a problem particularly though, because our
cultures overlapped at the time. They also had
jewellery and cave art. The issue would probably be
an unfamiliar form of neurodiversity, but parents deal
with that all the time, and parenthood is rarely
something one is prepared for. You just deal with the
new person without previous experience because
everyone is already unique. |
|
| |
23andme already has a tool to find other people with
high % Neanderthal admixture. Let 21st century
nature take its course for a dozen generations and
soon a Neanderthal population might reemerge
naturally. |
|
| |
Baked - see current occupant of Whitehouse. |
|
| |
//resurrecting extinct species is already widely discussed in
science and science fiction literature// |
|
| |
Half baked for a certainty, in general principles, yes. |
|
| |
//should be original to the poster// |
|
| |
Show me anywhere else where's it been suggested we
source our DNA building blocks to make a Neanderthal from
cabbages ;p |
|
| |
Perhaps I've been remiss in not showcassing the core idea as
clearly as I might. |
|
| |
Old fossil DNA is very broken up & damaged causing
difficulties in stitching it back together into something that
can be used by an egg cell (a la standard documented &
replicated cloning techniques) to produce a useful
organism. |
|
| |
& jigsaw the genome you want together from living DNA
extracted from extant animals & even plants. |
|
| |
I don't myself recall seeing that proposed as a technique
for
restoring extinct species in any prior art. |
|
| |
But I doubt I'm ahead of the curve & stand ready to be
corrected :) |
|
| |
//filled in the blanks from frogs// a phrase with up to unlimitless usability |
|
| |
I stand corrected, but I do think it might be fair to say I've
gone a bit further than
that with stitch it 'all' together from extant animals & plants? |
|
| |
//I wonder if dinosaurs taste like chicken// |
|
| |
Nope, like frogs, which don't actually taste entirely like
chicken, speaking from personal experience.. though, maybe
if
I'd cooked them first.. |
|
| |
There's problems with all this %similarity stuff. Simplistic at
best. I've never read what it's really even based upon. The
problem with generating a new & functional neandathal
genome starts with knowing quite a lot about it. Once you
have a good idea from several samples, you can start to be
confident about the primary sequence. Then you can start
to build epigenetic maps like has already been started. But
you need to be confident about the effects of time and
degradation on all of those. |
|
| |
They already have the neanderthal mitochondrial genome,
that's pretty interesting. A human cell with neanderthal
mitochondria is very possible and would be pretty
interesting to study if, crucially, the money is available. |
|
| |
I guess you could carefully assemble the neanderthal
genome, section by section, or, mutate the human one,
section by section, but it's a HUGE job. Far easier to wait
for custom synthesis to get to the point of whole
chromosomes. Still enormously tricky to get that primary
sequence all wound up and viable in a cell. There's 10 years
of work there before you go into the ethical nightmare of
getting a viable embryo. |
|
| |
Umm.. I was under the impression that we'd already
sequenced the full
genome not just the mitochondria [bs0]? |
|
| |
Have I fallen afoul of hyperbole? |
|
| |
Umm, nope <linky> <linky. |
|
| |
3 billion letters, the whole thing, &
that was 10 years ago, we got the mitochondria
sequenced
2 years b4 that. |
|
| |
Point is there's only a 0.3% disparity between it & the
human
genome so all we need to do is jigger that 0.3%. |
|
| |
20% of that 0.3% can be found spread around in the non-
african population. |
|
| |
So that leaves us 0.24% to locate in other extant animals &
plants (or synthesize from basic DNA 'letters') & splice in
with CRISPR. |
|
| |
Plus the mitochondria which I assume has similarly minor
differences to ours. |
|
| |
I think your data may be a little dated [bs0]? |
|
| |
Or did I misunderstand your thinking? |
|
| |
Oh sequencing, sure. We can sequence whole
genomes for a couple of thousand dollars now.
(Actually, I should check if it's exome or real whole
genomes) The problem is that 0.3% of 3 billion
bases is an awful lot of mutations, most of which
will be spread around randomly, many won't do
anything. The problem is doing all the edits,
CRISPR can do it, slowly. But it's maybe 10,000
separate edits and screens? That's a HUGE job. |
|
| |
No doubt, thought it would be, but very plausibly doable now. |
|
| |
//...But I doubt I'm ahead of the curve & stand ready to be corrected :)// |
|
| |
//Old fossil DNA is very broken up & damaged causing difficulties in stitching it back together into something that can be used by an egg cell (a
la standard documented & replicated cloning techniques) to produce a useful organism.
So don't try to do that.
Use living DNA instead & jigsaw the genome you want together from living DNA extracted from extant animals & even plants.
I don't myself recall seeing that proposed as a technique for restoring extinct species in any prior art.// |
|
| |
I think the main problem here is that the way you're imagining that people would go about such a task is very much not how anyone who was
working in the field would consider going about doing it. |
|
| |
I mean, saying that "Old fossil DNA is very broken up & damaged" and that this might cause "difficulties in stitching it back together" is a bit of
an understatement.
You know how some of the materials in old cars can be salvaged, and the first stage of this is putting it through a car shredder?
I suggest that it would be easier to rebuild the exact model of a car from the shreddings of several similar cars than to try to physically
reconstruct an ancient genome using only ancient DNA.
And if you want it to actually be the genome of a living cell... well, there's more issues there. And more if you want it to develop into an
organism. |
|
| |
//So with CRISPR gene editing all we have to do is find the gene sequences we want edit some human DNA & inject it into a hollowed out
ovum.// |
|
| |
That's not how CRISPR gene editing works actually. You've conflated a couple of things there.
CRISPR lets you make an edit to a DNA sequence using an RNA template. Not replace it at the scale of entire chromosomes. (It's also my
understanding that with the current state of the art, there's also a down-side in that there's a tendency to introduce off-target mutations,
which is obviously not desirable in a massive conversion process such as this.)
I think you're mixing this with the reproductive technology of "somatic cell nuclear transfer" - in which a cell nucleus is implanted into an
enucleated egg cell. Currently it's not a technique with a high success rate. |
|
| |
Obviously it would be theoretically possible to use these techniques sequentially, and a rather convoluted process involving these (and much
more) seems to me more plausible than trying to work up fossil DNA directly - but I hesitate to call it trivial.
A couple of years ago there was some furore when a Chinese group claimed to have successfully gene-edited twin human girls. (for information
search for the lead scientist involved "He Jiankui".) It's not clear whether this was actually true or not - papers were rejected on ethical
grounds. And the modification doesn't actually seem relevant to the medical case.
But it's clear that people have considered doing more significant germ-line modification - hell, I've mentioned it on here fairly recently. At the
point where gene-editing is ready for the big time, I hope we'll be willing to use it.
Although- I must confess I'm very interested in seeing how you get ethical approval for winding someone's children back to the stone age. |
|
| |
Mainly this is a bump, to solicit a reaction to the above. But also: |
|
| |
//There's problems with all this %similarity stuff. Simplistic at best. I've never read what it's really even
based upon.// |
|
| |
I think this should be said explicitly, and perhaps less diplomatically. |
|
| |
Have you seen pop-sci articles, or heard people claim something like ~'humans are 50% similar to bananas'~? |
|
| |
There was probably some research which did indeed report that similarity (or even, sometimes, identity).
But without details of the comparison, it's at best meaningless, and at worst misleading. |
|
| |
There's many things that could mean. Is it a comparison of proportion of somewhat-similar genes they have
in common? It it a comparison of the sequence of specific genes, or their cognate protein sequences? If so,
what was the comparison algorithm : was it pairwise or multi-alignment? How did it extract the sequences
which were compared? If you don't have at least approximate indications for these, there just isn't enough
context to understand the result. See the link I've added. |
|
| |
So basically, one current typical approach is to identify one or several genes present in all organisms of
interest. The more evolutionarily related the group, the larger the set could be. In the modern era, one can easily
extract genes from a draft genome, but historically matching regions could be isolated and sequenced.
However you've got them, you need to compare 'em somehow. If we're looking at rather unrelated organisms, the DNA
won't be that similar, so it makes sense to translate that to get the protein sequences. This increases
similarity in a couple of ways - introns are naturally discarded, and all codon variations are stripped. You may even
count different amino-acids as similar.
Nevertheless, some of the
sequence won't match at all - or at least, not convincingly - and these ranges may be discarded from the
analysis. How to deal with insertions or deletions? One approach is to ignore them! If you're doing a multiple
alignment, the more variable regions will probably end up totally clipped away.
Regardless, the bits which are left have an overall similarity, which can be gleefully reported in misleading
pop-sci articles without any of the boring technical details.
Of course this varies depending on which genes (or set of genes) you compare and the alignment algorithm
you use.
Also of course, whatever the value is isn't at all a measure of overall similarity, or of evolutionarily relevant
divergence, only of relatedness of the parts which have a chance of being similar. |
|
| |
A somewhat historically used approach was microarrays, but my experience seeing those in action wasn't
flattering, and as far as I care all such data is obsolete now, if it wasn't at the time of generation. |
|
| |
An even older technique was to hybridise DNA from two organisms and examine heteroduplexes by EM or so.
I think most groups doing this probably knew what they were doing and were quite careful because it
must've been a royal pain to do - but even so it probably generated a lot of gee-whizz pop-sci factorrhea. |
|
| |
Another approach is just to count genes which appear somewhat similar as the same, and see what
proportion are shared. Obviously this also has implementation details - but even disregarding those, if the
result is cited without mention of the method, you can achieve impressive levels of similarity. |
|
| |
//Mainly this is a bump, to solicit a reaction to the above// |
|
| |
[Remains otherwise silent] |
|
| |
[Loris], having read both your last two annos all I can say is
you apparently didn't really read what I wrote. |
|
| |
// easier to rebuild the exact
model of a car from the shreddings of several similar cars
than //// an ancient genome
using only ancient DNA // |
|
| |
// seems to me more plausible than trying to work up fossil
DNA directly // |
|
| |
I am well aware of the 'difficulties' which is why I proposed
a different approach entirely. |
|
| |
Yet your entire response appears to be predicated on the
idea that I was proposing that? |
|
| |
And that's just one of the mistakes you've made about what
I said. |
|
| |
But don't expect me to go through your two walls of text
piece
by piece for you, you'll just have
to
make do with that one example. |
|
| |
That's not what I said, try again. |
|
| |