5 responses to “Spaghetti Logic”

1. 

   brunilda

   November 15, 2004 at 6:55 pm

   Kate,

   Thanks for the posting. However, you and Malkin seem to have misread the editorial. The way I read it is that the ultimate goal is to develop targeted medicines that by-pass crude approximations like race. That’s all. The fact that race (ehtnicity is a better term, race applies to all humans) is a crude approximation stems from the fact that intra-group variability is still pretty high.
2. 

   Kate

   November 15, 2004 at 8:27 pm

   I read it twice.

   Though I’m a lay person, I have more than a passing familiarity with genetics. To call race a “crude approximation” may be accurate enough on a political level, but at the dna level, race can indeed be a very reliable indicator of risk. (See diabetes in North American Indian populations).

   Without putting to fine a point on it – the same argument used to be made about purebred dogs – that so few genes were involved in distinguishing between closed breed populations that there was no real definition of “breed” beyond the superficial phenotype. ie: that if you looked at the genome of a poodle and a bulldog, you would not be able to spot appreciable differences.

   Then a study was conducted into dna profiles of several different breeds. The researchers were able to identify breeds with 100 percent accuracy.

   It’s another case of the double standard of the “left”. These are the same people who state that religious beliefs have no place in deciding policy for stem cell research – yet they are religiously devoted to politically correct constructs about race – to the point of hoping that we can “move beyond” it. As though there is some unspoken higher purpose in doing so.
3. 

   brunilda

   November 15, 2004 at 10:30 pm

   Thanks Kate,

   I would like to have the references for the study on dogs you mention.
   Second, I was trying to locate the study the NYT refers to in order to compare the inter-group with the intra-group variance. Do you know where to find it?
   On the point about the political correctness of “race”: I don’t read the article as saying that such targeted medicines are wrong because they used the concept of race: it is actually complimenting them. It is also recognizing that more accurate assessments of genetics and the way genetic endowment relates to illness may allow us to target specific individuals and not only specific races. Saying that you can do better than race is not necessarily a double standard. And surely the validity of the argument can be examined without reference to who made it. Remove “crude” if you will: would you be satisfied by my rephrasing above?
4. 

   Kate

   November 16, 2004 at 12:09 am

   http://news.bbc.co.uk/2/hi/science/nature/3730169.stm

   A BBC article, that I believe, references the study. The original one came across one of our k9 genetics lists, too much work to look it up. 🙂

   As far as the other goes – I know James wrote about it a couple of weeks ago. Try searching the archives here, or just advance google until you find it.

   It’s all well and good to speculate about using definitive genetic testing on individuals to see who is predisposed to disease A or disease B. Until you factor in the costs involved, the number of potential tests, insurance issues… and the fact that many disease conditions are not caused by simple, single gene defects (or even defects at all) and that often, a disease broadly labelled under a single description can have several potential genetic origins. DNA research in dogs is proving this in spades – definitive gene tests in one breed are, more often than not, useless in others. If memory serves, there are 8 or 9 different genes identified as causitive for progressive retinal atrophy. Same clinical diagnosis, different gene defects. Some breeds have more than one type.

   And, the fact that this drug seems to be more effective on African Americans may not be related to a specific type of inherited heart disease at all, but to a multiplicity of genetic factors that function as catalysts – that were concentrated by natural selection to provide some advantage in the ancestors of modern day humans of African descent that are in and of themselves, quite “normal”.

   One of the problems in reporting of new discoveries in genetics is that so very few people understand how genes are or how they function in the first place – or how much of the knowledge of only a few years ago has been consigned to the dustbin. You cannot rely on what you learned in your high school biology any more. The textbooks are being revised almost daily now.
5. 

   brunilda

   November 16, 2004 at 2:04 pm

   Thanks for the links.

   So the point still is the perceived double standard held by the author of the NYT editorial.

   As Kate stated it, the NYT struggled to “remind readers that race is a concept with no basis

   in scientific fact, while encouraging the FDA to approve a heart medication developed

   specifically for African-American patients.”

   The last sentence of the first paragraph of that editorial states: “Thus it is a welcome

   sign of increasing national maturity that medical experts, both black and white, are now

   grappling openly with the issue of race as it applies to medical treatments.” So the second

   part of Kate’s contention, that the NYT was “encouraging the FDA to approve a heart

   medication developed specifically for African-American patients” seems to be true.

   The first part of Kate’s contribution states that the NYT wanted to “remind readers that

   race is a concept with no basis in scientific fact”. The NYT statement that comes closest to

   this is: “The chief drawback is that race is too superficial and subjective a concept,

   mostly based on skin color, to match up well with any underlying genetic or physiological

   differences that may affect how an individual responds to a disease or a drug treatment”. In

   particular, the concern is that “there is considerable genetic variability within any racial

   group, so it is likely that the new pill may fail some black patients, while white patients

   who could benefit may not get it because they don’t fit the racial profile.” Note that this

   seems to be saying “hey, let’s not forget there are white people out there too!”, not a

   classical left-wing motto. And the author goes on to conclude that more specifically

   targeted medicins, based on more accurate assessments of genetic differences are the ideal.

   Now, is there any reason for this caveat? Kate says that, in the case of dogs, “The

   researchers were able to identify breeds with 100 percent accuracy.” However, the same

   article reports the following words of the researcher: “It’s a much more striking

   difference than is seen among human populations that evolved on different continents.” Which

   leaves me wondering whether the dog argument applies or not. Are we justified in urging

   caution? We probably are, and I think Kate actually did conclude that in her last posting. Especially when the dog breeds experiment adresses only inter-group variability, not intra-group variability. We don’t know how different the different dogs in each breed are to one another. The key is how much of the variance is explained by variance within groups as opposed to between groups. Without hard data on which to perform tests, we simply don’t know. As Kate goes on to say, there may be a myriad factors involved in how people respond to illness and medicine.

   Was the editorial a reflection of a certain discomfort with the concept of “race”? Probably. It is not very easy to approach a subject that is so laden with history and with social inequality. For a lot of people, including me, extreme caution is needed with conclusions based on race, because, one, science is fallible, and two, we as humans have a history of racism, present and past, that is difficult to forget. In that sense, I totally understand the writer’s discomfort. I don’t necessarily think it is a double standard: I just think there is a genuine tug-and-pull here that is not ill-meant. This happens all the time when you do science (although the editorialist probably doesn’t do science). Science doesn’t happen in a vacuum; it happens in societies. If the conclusion of the study had been that people who eat cereal four times a day benefit more from the drug, I doubt we would be discussing this. And the main reason is because there is no stigma associated with eating or not eating cereal.

   In any case, this was a great discussion. I as a scientist, as a linguist in fact, grapple with this all the time. Especially if you do sociolinguistics and you know that your training tells you that Black English is as good a language as Standard English, but you know also that Standard English has to be taught because it is the dominant variety. Am I guilty of a double standard? I would like to believe no.
   Cheers,