Go read the Disclaimer again. I am not a doctor. This is not medical advice. Seriously.
I'm not going to attempt to cover the details of sexual reproduction here. Instead, I'll focus on some rarely mentioned aspects of DNA that don't get necessarily get covered in standard treatments.
There are two kinds of DNA: nuclear and mitochondrial. The nuclear DNA is what we generally think of as DNA, but we wouldn't be anywhere without mitochondrial DNA. Mitochondrial DNA is passed through the rest of the cell; it is passed from mother to child: pure matrilineal descent. Not all nuclear DNA is confined to the nucleus of the cell; bits float around in our bloodstream all the time.
Maternal DNA both within cells and outside of cells can be found in the fetuses blood stream and vice versa. The placenta is not the barrier it was once thought to be. Furthermore, we do not yet know what all that DNA is doing in either the mother or the baby; anyone who claims otherwise is theorizing in advance of the data.
In our society, we assume people belong in one of two gender categories. Reality is far more complex, and may be better understood in other cultures. The Intersex Society of North America is doing its best to raise awareness and prevent surgery on babies born with anatomy that someone decided is not standard. Intersex has a number of causes, some of which are developmental and others of which are genetic.
A developmental form of intersex involves prenatal exposure to androgens. During the 50s and 60s, progestin was given to some pregnant women to prevent miscarriage (it didn't work for that purpose). Fetuses with XX thus received the hormones more appropriate for developing male fetuses.
A genetic form of intersex is when an individual with XY chromosomes develops with female genitals and testes, due to insensitivity to androgen.
It was believed at one time that all the nuclear DNA in a given individual (human or amoebic, or otherwise) was identical, subject to mutation and the like. However, we now know that sometimes a person has more than one kind of nuclear DNA, a situation called chimerism and the subject of some fairly stupid crime dramas on TV. This can lead to some weird situations that call maternity into question, which one would otherwise expect to be impossible. It can also lead to some forms of intersex.
I have been completely unable to determine the rate of chimerism in the population at large. Because chimerism is more likely in multiple fetuses, and because multiple fetuses are more likely with older women and women undergoing fertility treatments, it is not clear that population frequency would adequately predict frequency in new pregnancies in the US, as the average age of the mother increases and fertility treatments are more common, both of which increase the likelihood of multiple fetuses.
The presence of fetal DNA in maternal blood was discovered recently, and immediately people started trying to use that DNA for early genetic screening of the fetus. Testing the cellular DNA required complicated preparation. Worse, it was discovered that fetal DNA within cells could persist in the maternal bloodstream for decades. Then fetal DNA was found in maternal plasma (outside of cells); that DNA is currently believed to clear from the maternal blood shortly after delivery (and presumably after miscarriage or abortion). Best of all, it is much simpler to prepare fetal DNA in maternal plasma. Many tests are being developed to exploit these discoveries. Most of them must pass FDA requirements for accuracy.
However, gender tests are not considered medical tests. They do not need extensive proof to be put on the market. Acu-Gen Biolab, a company in Lowell, MA, recently offered a five weeks gender test that works on dry maternal blood. Initial claims were for accuracy of 99.9% accuracy. It is unclear why the makers of the test allowed for a 1 in a 1000 error rate. It would allow for the more obvious forms of intersex, which occur at about 1 in 2000 births. However, a number of women who used this test are now suing because the test was incorrect and in many cases they have not received their guaranteed double-the-money-back. The scientist behind the test is accused of saying some unpleasant things to women who have claimed inaccurate results, saying their babies would have problems.
While forms of intersex that are apparent at birth occur at a rate of about 1 in 2000, the rate at which chromosomal gender differs from apparent gender is not known. Some believe that rate might be 1 in 90 or even higher. That might account for the inaccurate test results and the maker's remarks about future problems. On the other hand, the maker of the test might merely be incompetent or have suffered from a run of bad luck, or have sold so many tests that the complaints are the expected 1 in a 1000 errors. Without a lot more testing, however, we must assume that the accuracy of a gender test that looks at fetal DNA in maternal blood is unknown. There are numerous points at which errors might theoretically occur, and an unknown number of points at which errors might occur in practice.
It is worth noting that the maker of the test originally only tested for the presence of Y chromosomes and added a check for X chromosomes as of August 2005. Clearly, they were a little surprised at what they ran into. They also require that women using IVF with multiple implanted embryos wait until 7 weeks before attempting the test.
A General Discussion of Prenatal Testing
Copyright 2006 by Rebecca Allen.
Created March 1, 2006 Updated March 2, 2006