From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Fri Jan 24 2003 - 21:52:01 MST
On Fri, 24 Jan 2003, Nathanael Allison wrote:
> Nate writes,
(Oh joy Nate a substantive question that isn't about the stupid Iraq thing!)
Thank you, thank you, let me wash your feet to thank you.
> I have read they will be done mapping out the human genome this year.
Ok, lets discuss some terms. "Mapping" in classical genetics is the
process of assigning some nanoscale sequence to some macroscale location.
So for example, you can go to the "Online Mendelian Inheritance in Man"
database (otherwise known as OMIM), [http://www.ncbi.nlm.nih.gov/omim/]
And if you go to the "morbid map" and type in "Huntington Disease",
you will find that the gene for that disease maps to 4p16.3 (a specific
location on chromosone 4).
This URL should pull it up:
http://www.ncbi.nlm.nih.gov/htbin-post/Omim/getmorbid?start=0&term=Huntington+Disease
It is worth noting that for every specific human disease which is believed
to have a "genetic" cause, it is assigned an OMIM entry and as information
about the disease accumulates, it becomes robustly documented.
The OMIM number for Huntington Disease is 143100 and the online record
is here:
http://www.ncbi.nlm.nih.gov/entrez/dispomim.cgi?id=143100
I'll simply note that the online description of Huntington's Disease
is in the vicinity of half a megabyte (larger than some novels) and
the last time I seriously worked with OMIM (close to a decade ago)
there were about 8000 human genetic diseases cataloged (many fewer
of them were mapped then than are now).
So, the term "mapping" as dished out in the popular press is really
inaccurate (and I don't have a clue as to why it got promoted).
The human genome (in terms of knowing the layout of the chromosomes,
where some of the disease genes resided, etc.) was largely "mapped"
by the mid-1990's if not earlier.
What the press calls "mapping", people who understand this call
"sequencing" -- i.e. one has to read the precise genetic sequence
of the A's, C's, G's, and T's that make up the DNA of the genetic
blueprint of an organism. (So if one wants to be viewed as "informed"
one should lose the term "mapping".)
Now, the complete sequences for several of the smaller human chromosomes
(some of from 21-23 I think [higher numbered chromosomes are generally
much smaller and therefore easier to sequence]) were published several
years ago. So they can be viewed as a "done deal". A number of genome
centers are working through the remaining larger chromosomes to resolve
the "difficult" to sequence parts. (There are a number of reasons one
can encounter "difficult" parts -- but suffice to say they are only a
small fraction (perhaps 1-10%) of the overall sequence.)
By and large the sequence (and mapping) of the human genome is *done*.
That doesn't mean the locations (and genes responsible for) the 8000+
diseases in OMIM have been identified. A decade or so ago, I think
only a few hundred disease genes & locations were "mapped". Having the
complete sequence and a much more robust map is greatly accelerating
work in this area because it is much easier now to determine when
a sequence is "abnormal" and may therefore be the cause of a disease.
Identifying a mapped location, and then a "gene" responsible for a disease
greatly helps in determining what the function of an "unknown gene" is.
(This is a critical problem -- we do not know the functional purpose
of something like 30-50% of the genes we identify). Though it is a
rather macabre situation, identifying the purpose of genes through
the diseases they cause is greatly accelerating medical progress.
> How long will it take to find out all the different combinations of each gene
> once they finish?
This question isn't clear. Gene variants are known as "polymorphisms" ("many forms").
There is one gene that has multiple forms in the general human population -- ApoE which
has 3 variants (E2, E3 and E4) and those variants appear to play a critical role in ones
cholesterol levels and how long one lives. There appear to be a large number of polymorphisms
in neurotransmitter receptors and transporters -- so many that I can't easily keep it straight --
this is Anders field)
(see: http://www.dna.com/capabilitiesServices/capabilitiesServices.jsp?site=dna&link=cg_cnspanel.htm
When a gene is severely broken (i.e. it has a polymorphism that produces an observable
disease) it will show up in OMIM and will probably be subject to analysis within the
next decade (or perhaps the next two decades). When a gene is less severely broken
(people have a slightly lower sex drive or are somewhat less intelligent) it will
likely take much more time to determine whether there are genetic sources for these
conditions.
The question of "different combinations of each gene" isn't clear to me.
Humans probably have 30-40,000 genes and perhaps some more RNA which may or
may not be exerting some significant influence. With that number of factors,
I don't think one could try out all the different "combinations" with a
population *much* larger than exists in the world today. So if the question
is about trying out all the different combinations -- it ain't going to happen.
> and what each gene changes in comparison to it's counterparts?
For the reasons I cite above -- it isn't going to happen. The best we
are going to manage in the near future is to determine the effects of
a few key gene variants in a few critical situations (e.g. the effect
of Apo E2/E3/E4 on cholesterol and longevity).
> Also what about genes present in some people and not in others?
That is easy -- the URL I cite above (dna.com) seems to suggest that
we already have companies promoting services so you can perform
self-genetic diagnosis (otherwise known as "genotyping").
> I would like to know a good estimate, if the technology is currently
> available or when the technology to make these discoveries will probably be
> available, thanks.
The technology is available *now* that you could sequence parts of your genome
(or even your entire genome) if you have the resources to do so. The URLs
I cite show that it looks like you can get the information out of ones genome
if one has the financial resources. It will be moving in the direction
from "I have an IQ or SAT score of X" to "I have a dopamine DRD3 receptor of Y".
What isn't available "now" is the information that would determine (with a
high probability) that with a dopamine DRD3 receptor Y that you are the next
Steve Jobs or a sex crazed psycopath. (In contrast the APO E2/E3/E4 data
is probably solid enough that one could set life insurance rates based on it.)
If you have managed to get through this to this point you know more about
genomics than almost any college professor excepting those who can claim
"ownership" to a specific human chromosome (or those that have sequenced
the entire genome of some lesser organism).
So go out and kick some butt.
R.
This archive was generated by hypermail 2.1.5 : Sun Feb 02 2003 - 21:26:03 MST