reply to post by Jean Paul Zodeaux
Originally posted by Jean Paul Zodeaux
reply to post by Rising Against
I am not so sure that is an honest answer. It may be your honest belief, but to this date, no one has been able to find the "gay" gene within human
Originally posted by Jean Paul Zodeaux
I was compelled to clarify that your opinion in no way counts as scientific analysis, and since DNA is coming closer to being fully mapped by
scientists, unless they are able to point to a specific gene that would constitute being born with homosexuality, your opinion will remain that, an
Looking at DNA to find specific genes related to a certain behavioral or phenotype is a good way to find out what causes those phenotypes - but DNA
being fully mapped by scientists is simply not true. We may have the technology to read the entire string of DNA base pairs (GC, AT, etc.) - but as
one of the 46 chromosomes contains about 220 million base pairs, we're not quite near understanding all alleles (see for example the recent studies
on subjects like 'DNA methylation and Z-DNA formation as mediators of quantitative differences in the expression of alleles" by Rothenburg et al. ,
2001). In addition, it is much harder to find alleles interacting with each other then it is to find, for example, a single 'baldness-gene' .
When it comes to a comprehensive and extensive mapping of DNA, there is none.
More importantly, while one specific gene may not be responsible for a complex matter as one's sexuality, studying DNA is not the only way we have to
study genetic influence May I introduce to you: behavioral genetics
- basically the study of nature and nurture. The assumption central in this
field is that differences among individuals in a population are due both to genetic differences between people, and to differences in their
environmental experiences. A fair assumption to make, since we cannot explain all by just genes or environment. If this assumption is true, then there
is a certain correlation in the expression of genes between twins; between monozygotic (identical) twins, the difference is caused by environmental
influences (including interactions in the expression of genes), while between dizygotic (fraternal) twins, the differences have the same origins as
between normal siblings; both genetic and environmental influences. For MZ and DZ twins, correlations can be calculated for any trait; the difference
between those correlations (which itself is a measure of the differences within each group) shows us what the genetic influence is.
Let's say, there is a correlation of 99% of red hair among MZ twins (only 1 in 100 times, 1 of the twins has red hair but the other hasn't - all
percentages in this example are fictitious). Among DZ twins, we find a correlation of only 50%. Coincidentally, siblings share about 50% of their
genetic makeup. What can we conclude from this?
Obviously, MZ twins have 100% the same DNA (give or take some differential/erratic base pairs). The first thing we can conclude is that there is at
least 1% non-shared environmental influence, since the correlation is only 99%. The missing percent is apparently not influenced by genetic makeup
alone, but by the environment; in addition, whatever interaction there was between person and environment, it was only experienced by one twin
(subsequently leaving one twin with red hair, lowering the correlation of the whole group).
Furthermore, we find that among humans who share 100% of the DNA, the correlation of red hair is 99% - when they share 50%, the correlation is 50%.
This seems to imply that the 50% DNA that isn't shared by DZ twins, is responsible for 49% of the correlation (99%-50%). By that logic, having 100%
the same DNA would provide a correlation of only 88% (49% * 2). But that leaves us with a difference of 11% from the 99% we found for twins! That is
exactly what we need, since we need to account for genetics, non-shared environmental influences and shared environmental influences - the 11% is the
latter. In this example, we have found a genetic influence of 88% for red hair, with 12% environmental influence (of which 1% non-shared).
As with most methods, there are weaknesses and strengths; the main weakness is that genes may not be expressed through interaction with the
environment, which would lead us to conclude a lower genetic influence than possible (even though the percentages would accurately reflect the
influences at that point in time). The strength is simply of a statistical nature - the more subjects there are in a study, the more accurate the
percentages. If the methods of calculation seems crude to you, you're correct, and finally more and more researchers are starting to use more
advanced statistical techniques such as SEM (Structural Equation Modeling).
Long overdue now are the actual numbers for homosexuality, so with a little further ado - new articles require some kind of subscription to the
publisher, so I'll be linking to old data: according to Bailey and Pillard's 1991(!) paper "A genetic study of male sexual orientation
), 52% of male monozygotic twins share their homosexual preference; only 22%
of the dizygotic twins do so as well. This leaves us with 2*(52%-22%) = 60% genetic influence.
For women, 48% of MZ twins share their homosexuality, and 16% of the DZ twins did; apparently showing 2*(48%-16%) = 64% genetic influence.
Due to space limitations, I can't describe the more recent results of Zietsch et al.'s paper "Genetic factors predisposing to homosexuality may
increase mating success in heterosexuals
" (published in 2008 in Evolution and Human Behavior), but suffice it to say that using SEM, they were
unable to make a useful model (high goodness-of-fit) without genetic influence, while environmental influence could be removed while still retaining a
useful model. In other words, based on statistical analysis, it is impossible to make a valid model explaining sexual orientation without accounting
for genetic influence.
We may not know which parts of DNA influence sexual orientation, but it's clear they exist.