h a l f b a k e r yThe embarrassing drunkard uncle of invention.
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,
|
|
|
As we are able to entirely sequence more peoples'
genomes
we're able to find the parts that are the same across all
of
them, the parts that only represent minor physical
differences, and the parts that represent very unusual
mutations. I propose an organization whose purpose
would
be
to create a baseline human genome consisting of all
the
parts everyone has, a generic mix of individualized
traits,
and no significant mutations besides these.
This model could be used to repair DNA when that
technology reaches full fruition.
For example a DNA sample contaminated by a virus could
be returned to normal even if an adequate sample could
not be found. Likewise an aging person's DNA could be
reset to something before its undamaged state. Of
course this wouldn't represent the person's actual DNA
but the odds of a serious clash would be much lower.
1000 genome project
http://www.1000genomes.org/
[Voice, Mar 04 2012]
diverse to the core, for a reason
http://www.theatlan...fect-health/252795/
[swimswim, Mar 04 2012]
23AndMe
http://www.23andme.com
[theircompetitor, Mar 05 2012]
Scientists propose project to build synthetic human genome
http://uk.mobile.re...eNews/idUKKCN0YO2HA
[Voice, Jun 03 2016]
[link]
|
| |
For instance having a lost limb replaced by a biological generic one instead of a mechanical prosthetic, yes ? |
|
| |
The first problem is that there is so much
variation (much of it very common), and so little
ability to correlate sequence with phenotype
(yet), that we'd struggle to define a "generic"
genome. |
|
| |
Moreover, there's no guarantee that a "generic"
genome would be viable. It's quite likely that
many genomic variants work well only in specific
combinations and (perhaps surprisingly) just
combining the commonest variants would lead to
a non-viable human. |
|
| |
Next, we have no way (nor any way in sight) to
edit the genome of a human (or any other
animal), except at the single-cell stage. So, you
could in theory create a human with the "generic"
or "flawless" genome, but you couldn't do much
for an adult human. That said, we can do very
limited engineering on limited numbers of certain
types of cells, and maybe in 50 years time we'd be
able to edit the genome in all the cells of an
adult. |
|
| |
Finally, there's another bombshell waiting to go
off in the human genome. It's always assumed
that your genome is the same in all your cells
(barring accidental damage), but this is going to
turn out not to be the case. There's already
evidence for quite extensive variation between
the genomes of different tissues in the same
individual - probably caused by mutations during
early development. |
|
| |
Over the next five years, we will discover that this
variation is not only significant for some diseases,
but I believe we will also find some instances
where the genome is systematically "edited" in
some cell-types as a method of regulation.
Nature uses every other trick to regulate genes,
and she will surely use this one too. (It's well-
known in antibody-producing cells; but so far this
is considered to be the only instance.) |
|
| |
This somatic genomic variation will add an entire
new dimension to the genome, and will multiply
the problems of genetic engineering
exponentially. |
|
| |
Having said all that, the basic idea of finding
problematic mutations (inherited or acquired) and
fixing them is the essence of genomic therapies,
and is being actively pursued. |
|
| |
[ ], because it's an interesting and valid idea, but
also a very well-worn one. |
|
| |
//Over the next five years, we will discover// |
|
| |
Who are you? When did you come from? |
|
| |
//Who are you? When did you come from?// Just
trust me on this one. |
|
| |
//Moreover, there's no guarantee that a "generic" genome would be viable. It's quite likely that many genomic variants work well only in specific combinations and (perhaps surprisingly) just combining the commonest variants would lead to a non-viable human.// |
|
| |
Even though I do know something about who Max is and where he comes from, I think he's laying it on a little bit thick here. |
|
| |
While I don't doubt that there are incompatibilities between occasional rare alleles, or even between common variants in very isolated populations, I can't imagine that there are strong incompatibilities between the commonest variants of any gene in any combination. The selection pressure against would be strong. |
|
| |
Yes, fair point. On the other hand, the great
majority of pregnancies don't go very far beyond
fertilization. There are also many couples who
can't conceive for reasons unknown, and many
others who have to try an inordinately large
number of times to get it to work. I can't prove
it, but I believe that in the interplay of 27,462
genes there are some combinations that just don't
work, even though all the genes are fine in other
contexts. Selection during early development
just hides nature's blushes. |
|
| |
It's not so much about specific allele
incompatibilities (though those will exist). It's
more to do with the fact that the cell is a sort of
turbulent system, and there are only so many
ways in which it can settle down into a viable
state. |
|
| |
Though I base this on nothing but gut feeling, I think
[MaxB]'s statements may be a bit on the conservative side.
Human understanding and exploration of the fundamental
properties of our bodies and minds, our environment, and
our very existence are expanding at a rate unforseeable
just a generation ago. Things that amaze my parents, I
take for granted. Things that amaze me appear so
frequently that I, a fiction writer with an imagination that
knows no boundary, cannot begin to concieve what
wonders my unborn children will someday take for
granted. Therefore, [Max]'s rather understated revelations
concerning the human genome seem perfectly logical to
me. |
|
| |
The somatic variation stuff is real (there are
already some clues to it); the non-viability of a
"majority genome" is just a gut feeling. |
|
| |
And as regards my comments on the impossibility
of editing the genome in all the cells of an adult
human, well, one day... |
|
| |
Mitochondria might be a good place to start.
They've got a tiny genome that causes an
inordinate amount of trouble, and one can at
least conceive of the vaguest chance of a remote
possibility of "correcting" the mitochondrial
genomes in a large number of cells. |
|
| |
hmm... if there's no generic human genome, then our ancestors are different genomes of prehistoric apes. |
|
| |
//I can't imagine that there are strong
incompatibilities between the commonest
variants
of any gene in any combination.//
Such incompatabilities exist: the term to Google
is "double heterozygote" |
|
| |
//an interesting and valid idea, but also a very
well-worn one// This idea resembles Francis
Galton's multiply-
exposed photographic portraits purporting to
show
"typical"
physiognomy of various races. |
|
| |
//<me>I can't imagine that there are strong incompatibilities between the commonest variants of any gene in any combination.//
//<mouseposture>Such incompatabilities exist: the term to Google is "double heterozygote"// |
|
| |
To be honest I don't think I said what you think I said. |
|
| |
//It's not so much about specific allele incompatibilities (though those will exist). It's more to do with the fact that the cell is a sort of turbulent system, and there are only so many ways in which it can settle down into a viable state.// |
|
| |
I certainly don't disagree that there may be strange and random interplay between the numerous elements involved. However, my take on it is that genes have evolved to cope with the constant mixing which occurs. The genes which do well tend to be the ones which play nicely with others. |
|
| |
I am not a human geneticist, but my understanding is that humans are not even a particularly genetically diverse species, taken in the round.
I say in the round because not one but two recent reports reveal that people in some existing populations carry sequence from other human 'species'[1] (Neanderthals and Denisovans). Brilliant. Now if that's possible think what itmeans for the piffling intra-species variation (within Homo sapiens sapiens). |
|
| |
[1] The term species becomes a nebulous concept indeed. |
|
| |
//The genes which do well tend to be the ones
which play nicely with others.// Yes, that's a fair
point, and you may be right. |
|
| |
The Halfbakery always feels a bit weird when someone annotates an idea in such a way that suggests they know what they're talking about. |
|
| |
If you like, I can take over for a while. I have no fucking
clue what most of this recent stuff is about. |
|
| |
Well based on the TV observation that O- blood (or possibly + or maybe neutral... or another letter... anyways...) can be transfused to anybody, we can start there: we want the genes that make O- blood. |
|
| |
Now is there a DNA combination for an organ that won't be rejected by anybody (or antibody as the case may be :) ? or possibly a compromise where we can permanently "dumb down" a body's antibodies a bit to meet a generic'ish DNA skin graft or something. |
|
| |
I'm curious as to whether or not a person composed of entirely generic dna would look like Mitt Romney. |
|
| |
O-neg is the universal donor. O-pos can go to anyone but
an O-neg. My mother is O-neg, my father AB. Both
T.G.F.J. and I are O-pos. Neither of her parents are O-
anything. I don't know for certain, but this suggests to me
that the O blood type is recessive--very hard to breed for. |
|
| |
The Red Cross calls us 'type-O heros' (sic) and is constantly
hassling us to donate with letters and phone calls, which I
frequently do even though I find the moniker vaguely
insulting, as well as the 'Gallon Donor' certificates and
lapel pins they send me every so often. I don't see what's
so heroic about being born with a useful blood type. |
|
| |
[theircompetitor] that link is incredible. I knew it
was coming but I
didn't know it had arrived already. |
|
| |
//suggests to me that the O blood type is
recessive// |
|
| |
It is. "A" and "B" are two different antigens - I
think they're carbohydrate groups - stuck on the
surface of the red blood cells. The Rhesus factor
is another antigen. |
|
| |
So, an "O negative" donor simply has neither the A
nor B antigens, nor the Rhesus factor, meaning
that their blood won't be recognized as "foreign"
by anyone else's immune system. |
|
| |
Conversely, "AB positive" blood has all three
antigens (A, B and Rhesus). Therefore, if it goes
into someone who is not AB positive (for instance,
they're A positive), then at least one of the three
antigens (B, in this example) will be seen as
"foreign" and will provoke an immune attack. |
|
| |
If I recall correctly, having even a single copy of
the
gene for, say "A" will result in the A-antigen being
on your blood cells; likewise for "B" and Rhesus.
Therefore, the only way to make an "O negative"
person is for them to not inherit any of the A,B or
Rhesus genes from either parent. In other words,
O-negative is recessive at all three genes. |
|
| |
It's easier if you use capital letters to represent
the presence of the gene or of the antigen, and
lowercase to represent the absence of the gene or
of the antigen. In that case, genotypes AABBRR,
AaBBRR, AABbRR, AABBRr, AaBbRR, AaBbRR and
AaBbRr will
all be phenotypically "ABR" (AB positive in the
customary notation). The only genotype which
will produce an "O negative" person is aabbrr. |
|
| |
[Edit: a quick check on Wikipedia shows that there
are actually a bunch of different antigens in the
"Rhesus" group, but only one of the, RhesusD, is
usually important. |
|
| |
I'm also puzzled by something. Incompatibility of
A, B or Rhesus blood grouping can be bad in
transfusions. But only Rhesus (not the A/B
antigens) causes problems during childbirth, if the
baby is Rhesus positive and the mother is Rhesus
negative. I don't know why similar problems don't
arise if the baby is, say A and the mother is O or B. |
|
| |
The ABO blood group system isn't quite like that, Max; it's multiple alleles at the same locus. A and B are co-dominant, O is recessive.
So the possible genotypes are : AA, AO, BB, BO, AB and OO. The first two are phenotypically A, the next two are phenotypically B, the fifth is phenotypically AB and the last type O. |
|
| |
Regarding why ABO typing doesn't usually cause problems in pregnancy, apparently antibodies against these antigens are predominantly of a type which do not cross the placenta. Antibodies against the Rhesus D factor are of a different type which can cross the placenta. |
|
| |
//It's multiple alleles at the same locus// D'oh!
You're right, I am being a nidiot. |
|
| |