Field of Science

the Tamiflu scam

I read today that Tamiflu - the drug that supposedly helps get you better when you have the flu - is now available in ample supply, so Roche doesn't need to make any more. What a scam this is, all taking advantage of public hysteria over the "bird flu" threat (more precisely, the highly pathogenic H5N1 strain of avian influenza virus). Roche advertises Tamiflu on its own website as for both flu prevention and treatment - first of all, there is no evidence I know of that Tamiflu will prevent the flu. If you want to avoid it, get the flu shot, which is a vaccine, and is usually quite effective. Second, the best result you can get with Tamiflu is by taking it right after you get sick, and even then it just shortens the duration of flu symptoms by 1 or 2 days. Not very effective, right? By the time you have the flu, you're usually way too sick to go to your doctor and get a prescription - and you shouldn't even do that, because if you do you're likely to infect others. And finally, we don't have any good evidence that Tamiflu works at all against avian flu.

Meanwhile, though, Roche has got governments (including our own) stockpiling the stuff! And the government officials in charge proudly point to their efforts as if they were doing something about bird flu. What nonsense. If they really want to do something, we need to invest in - and approve - a better, more rapid way to make the vaccine. Right now we make it by growing the virus in chicken eggs - a 50-year-old method that is far too slow and wasteful. We could grow it in cell culture much more quickly and more flexibly (they already do this in Europe), but no company seems interested in trying to get a new method through the laborious FDA approval process. The US gov't ought to just fund this directly, and stop paying for stockpiles of Tamiflu.

UPDATE: Today (May 1) I read this headline: "Roche questions world´s flu pandemic readiness". Turns out that Roche is criticizing the WHO for failing to buy even more Tamiflu! Unbelievable. The spokesman for Roche said, in a press release, "... we believe there is still a long way to go...in terms of meeting treatment preparedness...that could dramatically reduce death rates." This is absurd - Tamiflu doesn't reduce death rates - there is no evidence at all of that - and furthermore, they don't even have evidence that it is particularly effective against avian flu. They're just trying to hype their product and increase sales by scaring people. This is the worst sort of corporate greed. The WHO already has 30 million capsules, and Roche says that governments have ordered an astounding 2.15 billion capsules, all because of fear of bird flu - but this profit windfall isn't enough for them. Their behavior is disgraceful.

the Human Microbiome Project

I spent Sunday evening and all day Monday at a workshop at NIH on the Human Microbiome Project - a new project, not started yet but probably going to start soon. The idea is to sequence the bacteria and other microbes (viruses and even small eukaryotes) that live on our bodies. We have these bugs all over us - in fact, as one of the scientists at the meeting reported, there are probably 10 to 100 times as many microbial cells on/in an average human as there are human cells! Don't worry, the human cells are much, much bigger, so by weight we're still mostly human.

But the point is that we don't really have much idea of what these bugs are that are hitching a ride on us. Some initial DNA sequencing projects - one of which I was involved in, sequencing DNA from the human gut - have found that there are thousands of species, and that the bacterial micro-environments are dramatically different in different areas of our bodies. The mouth has its own microbial environment, for example, which is very different from the skin. But there are big similarities between people - so if you look at the bacteria living in the gut of two different people, they have a lot in common.

It might turn out that many of these bacteria are "inherited" in a Lamarckian sense - that they are passed on from parents at some point during childhood, or maybe even acquired from others in our community. We need to do much more work to find out, and that's what the HMP will be about.

The workshop had some excellent talks by scientists who have taken an early lead in these investigations - Jeff Gordon from U. Indiana, David Relman from Stanford - as well as discussions from Francis Collins of NHGRI about how this fits into their sequencing program. It's clear we have an enormous amount to learn about the bugs on our bodies, and it's very likely that these bugs will turn out to have some interesting connections to our health - so I'm guessing that NIH will fund this in a fairly big way. Stay tuned.

By the way, my friend Jonathan Eisen blogged on this too.

chimps more diverse than humans?

I heard on the news yesterday that a new study discovered that chimps are "more evolved" than humans. It's too bad that the public will get that message, because it is really misleading. What the study found, by studying the genomes of several chimp populations, is that chimps have slightly more genes that appear to be evolving rapidly in response to their environment (this is called "positive selection"). In other words, this one study - whose methods are somewhat inaccurate to begin with, and which can only look at the parts of the genome that encode proteins - found about 225 genes in chimps that are evolving rapidly versus about 150 or so in humans. No surprise there - humans went through a population "bottleneck" in our relatively recent past, long after we diverged from chimps. A bottleneck occurs when the population gets very small, or when only a small number of individuals end up passing on their genes (which has the same effect) to subsequent generations. Anyway, the method didn't even attempt to look at the "noncoding" DNA, which is about 99% of our genome - it only looked at exons. So the study seems to be mostly an attempt to grab a headline, not much more.
But what the heck are we supposed to think a journalist means by "more evolved", anyway? I think the implication is that somehow they are superior to us, and in some ways they are - better at climbing trees, for example! - but this phrase is almost meaningless. Bacteria are "more evolved" too, if you think about it.
The study appeared in the Proceedings of the National Academy of Sciences, which makes the text freely available - so at least it's open access.

Free software and Richard Stallman

So Richard Stallman was in my office last week, and we talked about - what else? - free software. Stallman is famous as the founder of the Free Software Foundation, and perhaps even more famous as the inventor of the GNU software that forms the basis of most of our Unix systems. (He also wrote emacs, my favorite text editor.)
Well, he's quite a character. First of all, don't use the words "open source", "intellectual property," or even "Linux" in his presence, not if you want to finish a sentence. He'll interrupt you to tell you that software must be "free", and no other word will do, and that "intellectual property" is a muddy term that conflates copyrights and patents, and that "Linux" should be "GNU plus Linux" because of all the GNU software that comes wrapped in Linux distributions. I enjoyed talking to him because I agree that software should be free, but he is incredibly dogmatic about it. I don't know if he convinces people who don't already agree with him, but at least he's very up front about his agenda.
He gave a talk in my department (U. Maryland Computer Science) afterwards, and if you're interested, his talks are on his own website.

Free the data

I've been working on the flu virus for several years now, and some of us have been trying hard to get the flu community to share data, but it's an uphill battle. We started a project in 2003 (when I was at TIGR) to sequencing 1000's of influenza genomes and release all the data immediately - so that the entire scientific community can benefit. We've done that, and we've published letters in Nature urging others to do the same, but most of the leading scientists still refuse. They prefer to sit on their data and milk it for whatever results they can discover until all their papers are published, and only then will they release data.
Meanwhile the flu continues to spread around the world and we aren't really keeping ahead of it. Genome data - human, bacterial, viral, whatever - should be released immediately to have the greatest benefit. Any public funding agencies should insist that work funded by them include data release.
I have similar opinions about making our publications free, and I'll try to put some of them in my next post.