Can DNA analysis detect race, gender or breed?

I’m not a microbiologist so I’m not looking for technical date. I’m just curious as to what DNA analysis can tell us.

Can it detect racial makeup. I could see affirmative action programs using this to detect fraud?

Can it indicate the sex?

In the case of a dog, for instance, could it tell the breed of the dog?

I tried a search and came up with nothing.

IANAMicrobiologist either, but obviously you don’t need a complex DNA test to determine the sex of the subject. Just look for the presence or absence of the ‘Y’ chromosome.

Off the top of my head, maybe, yes, and idunno.

Sex is definitely determinable, and this is done routinely by chromosome examination; the presence or absence of a Y chromosome should also show up very clearly in the techniques used for DNA testing.

Race is a harder question. While it’s been shown that differences show up on DNA tests between different racial populations, race is genetically complex, and it’s not genetically defined at all. Thus, while DNA tests may be able to roughly categorize some people by race (i.e., “this blood came from a person of at least partially West African heritage”), they’re very unlikely to ever do so for enough people to be very useful. They certainly won’t be able to identify the ethnic identification of people who have some choice in the matter.

Dog breeds: I suspect the case is the same as for human races, except that the populations may not be distinct enough to yield any result other than a big blur. I doubt that anyone has done a lot of testing on this, though I may be wrong. Certainly DNA testing has been used to settle pedigree issues, but that’s different.

BTW: this isn’t microbiology.

Off the top of my head, maybe, yes, and idunno.

Sex is definitely determinable, and this is done routinely by chromosome examination; the presence or absence of a Y chromosome should also show up very clearly in the techniques used for DNA testing.

Race is a harder question. While it’s been shown that differences show up on DNA tests between different racial populations, race is genetically complex, and it’s not genetically defined at all. Thus, while DNA tests may be able to roughly categorize some people by race (i.e., “this blood came from a person of at least partially West African heritage”), they’re very unlikely to ever do so for enough people to be very useful. They certainly won’t be able to identify the ethnic identification of people who have some choice in the matter.

Dog breeds: I suspect the case is the same as for human races, except that the populations may not be distinct enough to yield any result other than a big blur. I doubt that anyone has done a lot of testing on this, though I may be wrong. Certainly DNA testing has been used to settle pedigree issues, but that’s different.

BTW: this isn’t microbiology.

Chromosomal sex is easy to determine. However, this does not necessarily mean that gender is easy to determine, since gender and chromosomal sex may not necessarily match, especially since gender is as much a social as “biological” phenomenon. The same can be said for “race”. When does “white” begin and end? Very few alleles (I could possibly say NO alleles) are 100% restricted to a single “race”. In any case “race” as it is used socially is a meaningless concept biologically.

Many dog breeds are extremely homozygous (and their populations isolated), due to linebreeding and inbreeding to keep and accentuate certain traits. So it is possible in this case for DNA to detect which is the most probable breed of the dog. Of course, you’ll need a template first, a database of the most likely alleles for given breeds.

Among humans, there is no test for “race.” The closest that anyone has come is to identify a constellation of markers that may correlate with a higher prevalence of some series of markers by continent. (And even then, the identification is done by comparison of probability, rather than the presence of any single actual marker, as all the markers are found in humans from all continents.) Noah Rosenberg et al. published this study in Science: Vol 298 20 December, 2002 (file requires Adobe Acrobat).

The study that discovered this used “core populations,” so it is also possible that if the study is expanded to more populations with less clearly defined coherence, it may break down (or it may not).

As to Affirmative Action testing, it would probably be useless, anyway. There was a concerted effort in several states (most notoriously, Virginia) in the early part of the 20th century to classify people as “colored” using the “one drop” rule, so a person with substantially “white” ancestry might, indeed, have grown up in the black culture, self-identify as black, and, depending on the local conditions and the knowledge of that person’s neighbors of his or her “race,” have been treated as black with all the discrimination associated with that. In addition, the cultural identification “Hispanic” has no “racial” component, so no test will identify pretenders to that status.

They don’t use DNA to determine who is who or what is what. They compare it to other samples.

As it’s currently used, though, DNA testing relies on a handful of fairly easy-to-find markers and gathering enough information to identify a specific person[sup]*[/sup] (as for a rape prosecution) requires far less than one percent of one percent the total sample. It will be some time, during which computers will have to become incredibly faster (maybe becoming organic themselves) before we can take a sample and get an instant breakdown of a person’s genetic legacy, Gattaca-style. I don’t see it being used anytime soon for something as mundane as affirmative action tests. It would be prohibitively expensive.

[sup]*[/sup] Strictly speaking, a DNA test doesn’t really identify you, but it eliminates the 99.999999% of humanity that isn’t you.