« Oh so sinister! | Gene Expression Front Page | Merry Old England »
January 10, 2003

Some old time Gene Expression

Godless Capitalist, Gene Expressor emeritus, forwarded me this document. Select the bionculars in the menu, type (the "find" feature for Acrobat Reader) "QTL" and read the article titled A winning combination under the heading MULTIFACTORIAL GENETICS. Even if you don't have a technical background, the gist is pretty obvious. Here is a snippet:


A QTL study would typically begin by identifying two populations that differ in the quantitative trait of interest, mixing the two genomes by genetic crossing and then looking for genomic regions that segregate with differences in the quantitative phenotype.

Here is Godless' text from the e-mail:

the application to humans is conceptually trivial, somewhat technically challenging, but doable within a year or two with a lot of money. there's one obvious quantitative trait to look at too...

Posted by razib at 11:53 PM




Oh how I miss him! If he were here I would ask him how he would arrange, in humans, the enormous number of meioses needed for this analysis. But alas. Call me back when they've found the genes for left- vs. right-handedness, and then we can talk about intelligence genes.

Posted by: Charlie Murtaugh at January 11, 2003 05:25 AM


humans ain't fruitflies-so true....

Posted by: razib at January 11, 2003 12:44 PM


"mixing the two genomes by genetic crossing." That seems to call for an actual experiment with controlled cross-breeding instead of an after-the-fact study of a mixed population. In fruit flies, it just takes several jars and a few days. Try it in humans, and they'll really call you a Nazi...

Posted by: markm at January 11, 2003 07:25 PM


well-i'm assuming that godless is taking about human populations that are "mixed" already out there. yeah, it doesn't map 1:1-but the theoretical principle is established, and we can go from there....

Posted by: razib at January 11, 2003 08:24 PM