[extropy-chat] Eumemics

[Robert J. Bradbury]

On Thu, 8 Jan 2004, Adrian Tymes wrote:

> That said, this could be a real issue in the very near term.

Yep.

> All that's missing is a good, publically
> available map of genes to approximate IQ, which
> doesn't seem that hard to create.

Its harder than you might think if there are many genes
(say greater than a dozen) and they have low penetrance
(say are responsible for only 5-10% of the trait each).
These are the characteristics associated with diabetes
and heart disease predispositions and its been rather
difficult to produce a complete picture in spite of many
people working on it.

I suspect Decode, working with the Icelandic dataset
or the company that is working with the Estonian (?)
dataset may be the first to begin to unravel this
because they have the largest number of samples to
work with (and don't have the problems a U.S.
University would probably run into).

> A group dedicated
> to the necessary studies could possibly publish a
> fairly reliable one (with a disclaimer that it only
> covers the genetic component of IQ, which is not 100%
> of IQ - partly as a cover in case one of the resulting
> children decides to destroy their own potential)
> within two or three years.

Sooner than you think... :-)  [I just love Google...]
http://www.hum-molgen.de/bb/Forum9/HTML/000052.html

And then there is the work at King's College:
http://www.ncbi.nlm.nih.gov:80/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=11838529&dopt=Abstract

And even Pennsylvania State University has an IQ project:
http://www.ncbi.nlm.nih.gov:80/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=8024528&dopt=Abstract&itool=iconabstr

And then of course there is this:
http://www.ncbi.nlm.nih.gov:80/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=12556904&dopt=Abstract
which finds 4 loci and 3 specific genes they can identify.

Related abstracts also report negative findings for association
with the dopamine receptor and several other genes.

> It would have to be a private group, since no current scientifically
> trustworthy government could get away with such an effort politically,
> but the resources required do not seem to exceed several million dollars.

It would be interesting to research who paid for the studies
cited above.  One thing is for sure -- people with below average
IQs probably cost governments money (extra educational requirements,
perhaps a greater possibility for becoming criminals, etc.)

[I'm not stating things as facts but as reasonable possibilities.]

> But the map isn't available yet, so this selection could not be
> done with confidence today in most cases.

Well 5+ positive polymorphisms and several negative polymorphisms
certainly gives you a start.

Robert

P.S.  I think the Extro list needs a semi-AI filter that runs
all "claims" through google to determine whether or not there
is resonable data to suggest that you might want to be very
careful about the claims...  (No offense Adrian... :-))