James and the Giant Corn Genetics: Studying the Source Code of Nature

March 28, 2008

General Election Strength

ClintonvsObama

If anyone is following the race for the democratic nomination for president, one topic that gets discussed ad nauseum is whether Barack Obama or Hillary Clinton has a better chance of winning against McCain in the fall. The problem with such arguments is that there are different roads to victory and and the two candidates are strong in different parts of the country and weak in different parts of the country.

In order to better visualize this I’ve dusted off my knowledge of Perl to generate a map of the states in which each candidate has an advantage. Dark blue states are ones where Obama does ≥ 10% better than Clinton relative to McCain, medium blue are ones where he does ≥ 5%, and light blue ones are where his advantage is 1-5%. States where Clinton has a relative advantage are marked in green, and coded using the same dark ≥ 10%, medium ≥ 5%, light 1-5% system.*

*All calculations based on the most recent poll in each state, and using date pulled from a site I highly recommend Electoral Vote.

Things to keep in mind:

1. A relative advantage doesn’t equal a win. (For example Obama is 27% closer to McCain in Utah than Clinton is, but that still means polls show McCain winning 50%-39%)

2. Likewise a relative disadvantage doesn’t equal a loss. (Clinton has a 12% advantage in Massachusetts, Obama an 18% advantage in Illinois, but the democratic nominee would certainly carry both states.)

3. The margin of error on most polls in 3-5% so the lightest blue and lightest green states are ones where neither candidate has a significant advantage. If I did this map again in a week, after new polls had come out, a lot of those states might change colors.

March 1, 2008

Of Genetics and Bears

Filed under: research stories — Tags: , , , — James @ 2:05 pm

One of John McCain’s signature issues is the wastefulness of government spending and one of the issues he’s getting a lot of play with is a government funded study of the genetics of bears in Montana. A description like “the genetics of bears” leaves a lot to the imagination so I looked up what the study actually entailed. They’re using wire traps to collect hair from grizzly and black bears in the wild and then using the samples to study the population structure. From the article I found in scientific American, it looks like they’re only looking at species (grizzly vs. black), gender, and number of individuals in the population. I can see why this is interesting from an ecological perspective, but I thought they were going to be doing deeper genetic analysis. 

Congress gave 4.8 million dollars to the bear genetics study, though McCain says 3 million in his speeches. With that much money, and the rapidly dropping cost of 454 and solexa technologies, a grant that size should have made it possible to shotgun sequence DNA from a number of individuals generating gigabases of data. Assembling the genome of the grizzly bear probably wouldn’t have been feasible within the 4.8 million dollar budget budget, but the data set generated could have been used for all sorts of studies. Looking at how much genetic diversity is present in different sub-populations. Looking for previous bottlenecks in grizzly bear populations. Identifying alleles of genes under positive selective pressure in the population. Testing if one set of alleles has been under selective pressure in the time frame since humans crossed the land bridge into the Americas, and a different set was under selective pressure previously. Alternatively, after mapping a number of single nucleotide polymorphisms (Spots in the genetic code occupied by different genetic bases (A, T, C and G) in different individuals, it should have been possible to develop a quick and (relatively) cheap test to identify different individuals. All the estimates I was able to find put the grizzly bear population in Montana at 1000 or less, meaning the cost per individual of this population count was over 5,000 dollars. My girlfriend informs me there are several ways to estimate a population cheaply and efficiently, such as recapture percentage.

My point here is that 4.8 million to study the population genetics and genomics of bears is completely justified, but there are probably cheaper ways to do a simple population census. Just to be clear, given that bears are an endangered species in Montana, there’s no question that a good population census was needed. McCain, however, sounds like he doesn’t think any study of bears is worthwhile. “I don’t know if it was a paternity issue or criminal, but it was a waste of money.” I guess funding genetic research into wildlife is wasteful by definition? Or maybe he means any genetic research except human genetics?

McCain seems opposed to basic research, and Hillary Clinton and Obama mention science as little as possible one way or the other. I’ve decided the reason I like Al Gore is that he actually comes across as smart and interested in research, but interestingly enough he only let that aspect of his personality emerge after he gave up on being president. What it comes down to is that the sort of person I wish was running this country is by definition unelectable.

Powered by WordPress