Protein folding is fun and games, and more

Despite the obvious advantage of having humongous amounts of computing power at one’s disposal, it seems that some problems’ solutions yet rely on humans’ puzzle-solving machinery, and our ability to recognize patterns that elude our hard drives. How our neural circuitry takes the leaps of logic that it does is another question, but since the capability is available, the win-win situation should be capitalized on–you get the satisfaction of solving a puzzle, and scientists get a candidate folded structure for a protein, for which the only available structural information is the primary sequence.

“Gamers unravel the secret life of protein,” John Bohannon (Wired Science)

From inside the beltway: watching Waxman et al working the fields

This summer in DC, we live in interesting times. Not only are we unusually likely to hear gunshots on the national mall, or fear imminent and deadly collisions on the Metro – oh no. The greatest fears, the looming things that fill my life as an intern with fear and trembling, come from the throats of House Agriculture Committee members, the ominous lineup at a hearing, the writings of renowned Princeton professors. Environmentalists look at the House climate bill, its massive, morning-of-vote manager’s amendment, Senate stirrings, and “aspirational” G8 emission reduction targets in awe. I have found myself open-mouthed at my computer screen and the television. “They must know what that means,” I’ll say, until the next shortfall, the next halfhearted compromise – “or don’t they?”

All around my ideological corner, we are alarmed. But for policymakers all around this fine city, global climate change seems – at most – just a wave to ride.

My area of expertise when it comes to wave-riders lies in the agriculture sector, where the waves are quite crowded. Biotech, ethanol, organic, you name it – if it’s ag, it wants a piece of the climate legislation pie*. And boy, is ag ever eating well.

Ag’s slice of Waxman-Markey (or HR 2454, or just “the House cap-and-trade bill”) comes in diverse flavors. First, unlike practically every other sector of the economy, agriculture’s greenhouse gas emissions – which amount to over 7% of national emissions (see EPA GHG inventory) – are not to be reported, regulated, or capped. This, on its own, is an apparently arbitrary** boon to agribusiness that many environmentalists likely find reproachable.

Additionally***, rural electrical cooperatives would receive a considerable portion of cap-and-trade permits for free under the House bill – understandable, given that co-ops are nonprofit organizations serving vast swaths of America.

But wait! Most of ag’s pie slice consists of juicy offsets. If an agriculturalist did; happen to reduce his or her emissions under Waxman-Markey, he or she could have those reductions verified as offsets and sell them to businesses in capped sectors in place of emissions allowances. In other words, ag can make money doing what other sectors have to spend money doing.****

What constitutes an offset? On the official list, introduced in an amendment by House Agriculture Committee Chairman Collin Peterson (D-MN), are plenty of organic practices, such as cover cropping and reduced fertilizer application, along with more expensive no-till methods and waste digesters. Some claims are legitimate and some are questionable, but so far, regardless of scientific validity, ag has gotten practically everything it’s asked for.

I cordially invite you to take a look, for instance, at recent and past research on the usefulness of no-till agriculture to the end of carbon sequestration in soils. You can find an interesting discussion of that controversy over on the Climate Progress blog. The current, very erasable bottom line is that we don’t really know if no-till does any good so far as climate is concerned – but there it is on the list, emblazoned by Chairman Peterson because, yea verily, this pie is tasty.

All that said, inside the beltway, we are quite aware that health care is the issue of the day. The Senate will not finish markups on its climate and energy legislation until September. With whatever additional time is allotted – I hazard that additional time will be allotted – the legislation will hopefully come to a vote and become law (or not) in time for the Meeting of the Parties to the Kyoto Protocol in Copenhagen this December. There’s still a lot of room for change to occur – probably negative change, from an enviro’s perspective, since the Senate ag bloc is even stronger than that in the House. The slice will grow larger or, if you prefer, the wave-riding will not cease.

But Science in Society is about the interface between science and policy – in essence, about trying to collapse the difference between what is usual, probable, and political and what is right. So I encourage you to do your own research on the provisions of Waxman-Markey, and on the activities of the Senate, and get in touch with your congresspeople accordingly. This is your country, your planet, and – let’s face it – your lifetime that will see the impacts of this legislation. Don’t let lobbyists or the party line gobble it up.

————————–

*Please excuse my abominable mixed metaphor. They happen to the best of us.

**Of course, nothing is arbitrary. Chairman Peterson assembled a voting bloc containing enough farm state Democrats to kill Waxman-Markey if concessions to ag were not made. Well, shucks – politics.

***There is yet another flavor of ag handout pie, and that flavor tastes of corn starch and burnt rainforests. Yes indeed, everyone’s favorite climate controversy: bioenergy. It is a topic deserving of its own post, so I will tell you about it later. Promise.

****If it sounds like I’m needlessly ragging on hapless small farmers, please consider that the vast majority of America’s agricultural products are made by large-scale, industrial agribusiness. Very few quaint red farmhouses have been harmed by my enviro-vitriol.

A quantitative biologist clueless about quants nonetheless muses about them

An obligatory link to an article about quants, now the dust has settled a bit: Recipe for Disaster: The Formula That Killed Wall Street, in ways of a delayed follow-up to Sarah’s previous post.

Also, a slight update, and a past opinion piece.

And finally, another word from Wired:  The Plight of the Quants.

This phenomenon was something I was vaguely aware of—the serious study of economics is mathematics-heavy, and mathematics is a wonderfully translatable tool, so the fact that Wall Street started recruiting physicists and mathematicians doesn’t seem surprising, at least to an outsider. However, how does their skill set differ from that of pure mathematicians on one end and economists on the other? I’ve occasioned glances into the conceptual rifts between physicists and mathematicians (when Bill Bialek was lecturing for Integrated in my freshman year, he would often apologetically, but bemusedly, point out that particular approximations are anathema to purists—this turned out to be a recurrent theme throughout the first two years of Integrated).

This brings me to a second personal observation, and that is the parallel between quants’ emergent influence and the movement of physicists to biology in the last half century. Our biology departments (all two of them) are filled with those who want to use quantitative methods to study problems, whether it’s because they allow you to zoom in on areas of interest instead of searching an entire space, attach probabilities to certain models and hypotheses, generalize specific problems so that you can see a continuum of similar problems, predict outcomes., etc. Why was it that they wanted to study biological problems? Is it because life poses a different kind of challenge, just as unexplored and mysterious as our universe at large, and tickles their fancy for problems with deep-reaching consequences? Because living organisms are able to turn normal physical laws and properties on their head, rendering them seemingly paradoxical and unrecognizable, and therefore begging explanation? Because people gradually began to realize that biological problems can be described by the same differential equations that describe other physical phenomena? I say these things partially to poke fun at empty dreamers, partially because I suspect people are much more whimsical in their career choices than I expect them to be, and partially because, in all honesty, there is something to be said about choices that people often claim were random and whimsical. There were probably plenty of objective reasons behind biologists’ realization that they needed to turn to a different toolbox, but I wonder more about the psychology of those that were involved in the paradigm shift.

When I was at UPenn over the summer, I heard about a few string theorists (including the professor I worked for) who had turned 180° and started working on information theory in biological systems. This is entirely different from the streamlined university educations that we have today that try (with certain success) to churn out individuals who have quantitative biological backgrounds, and start work on biological problems from the get-go. These programs attract a different kind of people, so I wonder how quantitative biology’s culture will change in 10 years, if at all.

The interesting discrepancy between physicists’ induction into biology and the article’s description of quants’ role in financial institutions is strange to someone who’s so familiar with the day-to-day physics envy that pervades the natural sciences (I’m a molecular biology major), even if it is often self-referential and invoked humorously. Much of the anger directed at quants by traders, businesspeople, etc. just seems to be part of the economic blame game, rather than some existing historical rivalry, and the public’s anger is sure as hell one of those “Let’s torch the unknown” kind of gig. I’m sure that if the economy were doing fine, no one would even hear about quants, who would anonymously continue their financial research. It was pointed out by one of the interviewees in the primary Wired article that science isn’t about making money (nor is it about never making mistakes)—there is a pretty big difference between people who go into financial engineering from the get-go, and people who end up there because they liked math and ended up in applications.

And even then, being a college student, you see where adults get their start and you can manage to be a bit more forgiving. Sure, it makes you wonder how many random (and sometimes anguished) decisions ended up shaping our lives irreversibly (I’m reminded of a friend who swung over to English from astrophysics). It also reminds you of how immature people start out in a field, and then proceed to grow up, becoming empathetic and worldly—but stay with the occupation from a previous life’s choice. And yes, I am making the generous assumption that most people grow up. My point is that many people like making money, and it makes me feel uncomfortable when people who have been making an honest living have had their careers ridiculed and scorned by other questionably self-righteous adults just because they were unfortunate enough to be stuck in the 21^st century with greedy management obsessed with short-term goals. I mean, how do you ever pinpoint the source of the problem? It’s easy to paint in broad strokes and act as if the cause was conveniently captured by a particular group of people in particular window in time. If there’s anything good that has come out of this at Princeton, it’s the heightened interest in charity work and taking time off before settling down with a job (according to a survey of the student body conducted by the student government, results available here). I don’t think it’s particularly great that all the future problem-solvers are scared, because it’s going to be difficult to solve an economic problem if all the young people are running in the opposite direction (if Wall Street is even interested in hiring anyone these days), but at the same time, forcing goal-oriented people to slow down and grow up a little more before taking on a full-time life commitment is not a bad thing at all.

End-of-term fun

For anyone with a final coming up, or with a hankering for some Tom Lehrer, a rare recording (and lyrics) from a musical comedy he wrote in the ’50s while teaching at a certain community college up in Massachusetts.  A fun listen for review session season:

Ha, he asks if there are any questions.
Holy smoke have I got questions!
I’ve got a ton, and every one,
Would take him half a day to do.
But I don’t really want to stay here
Since he’s said all he has to say here
But it’s agreed that I shall need
Much more than luck on the examination.
And so I think I’ll let him say goodbye
I guess that he is as relieved as I
Goodbye, goodbye,
Thank God the course is over now, goodbye

A performance in 1951

Trivia: Tom Lehrer, of course, became a very successful satirical songwriter — one of the two or three most widely known during the 50s, 60s, and early 70s. Two of the other cast members were advisors to U.S. presidents, one became a photographer for Time and Life magazines, and three others became highly-regarded professors at Tulane, Case Western, and UC Davis. The recording of the show was made by a young professor, Norman Ramsey, who went on to win the Nobel Prize.

Google on H1N1 swine flu

For your datahead enjoyment, a user has added a layer to Google Maps that lets you track cases of swine flu as they are reported.  We are still very much at the point of discovering preexisting cases of swine flu, so this will be a lagging indicator, but it’s still fun (and a little unsettling) to see how many cases of the disease have been spotted in the vicinity of Princeton.

A more utilitarian Google widget is Flu Trends, which tracks common search terms that, according to a Nature paper published by the company correlate well with actual disease levels.  (These terms include searches for symptoms that are related to flu; as Google points out, you get more allergy searches during allergy season and more sunburn searches during the summer.)  This can actually predict disease loads two weeks before health authorities announce new outbreaks.

Considering the hubbub about H1N1, it may surprise you that Flu Trends shows that influenza levels in the US are low and steady.  In the media blitz as new cases are discovered, it’s important to keep in mind that the normal seasonal flu causes 36,000 deaths a year.  While swine flu is newsworthy because of its potential to mushroom into a full-blown pandemic, in absolute numbers it is still a minor player in the disease world.

While the disease cannot be acquired by eating pork, pig farmers are naturally concerned that calling H1N1 “swine flu” will hurt sales.  However, the disease did originate in pigs, recombining with avian and human flu viruses, so it is unlikely that the industry’s calls for it to be redubbed will be heeded.  (“Mexican flu” was one suggestion, and I can think of at least one group that would be riled up over the name.)

Update: A remarkable new website allows you to tell whether or not you are suffering from swine flu.  This represents a significant step forward for online diagnostics.  Get yourself checked out here!

Flu intervention: then and now

Policemen in masks, San Francisco, 1918

In our earlier swine-flu related post, we mentioned that early intervention may well make the difference between an isolated outbreak and a deadly repeat of the 1918 Spanish flu.  Just how important is starting countermeasures early, and what kind of interventions work?  The tragedy of the Spanish flu provides a natural laboratory for public health measures, as cities throughout the US differed both in scale and timing of their interventions.

Medical science in 1918 was still getting on its feet.  The majority of older physicians of the time were not educated under the scientific regimen of the Flexnerian revolution.  The leading bacteriologists of the day mistakenly believed that influenza was a bacterial disease, and it was not until 1943 when it was recognized that a virus was responsible. As a result, medical intervention in the pandemic was of questionable value, not least because most of the best doctors had been drafted to serve in the military for WWI.

However, nonmedical interventions were also employed.  These included quarantines, isolation of the sick in makeshift wards, closure of public gathering places such as churches and schools.  A recent study examined the effects of timing and duration of these measures, with the major findings summarized in two graphs:

C: Mortality vs. time to intervention. D: Mortality vs. length of intervention

The study examined the experience of 23 cities in implementing various public health measures, and measured the impact of response time and duration of intervention.  They found that quick action (as measured by when flu cases rose to double the baseline number of cases) had a strong correlation with reduced mortality, and that maintaining the measures was important to keep the disease from spreading.

St. Louis, for example, closed schools and canceled public gatherings early, and maintained quarantines for over ten weeks, leading to a significantly lower mortality rate.  However, not all cities were as proactive; the median duration of these interventions was only four weeks, insufficient to protect the population.  Some cities were even counterproductive: Philadelphia hosted a military parade to promote war bonds, over the objections of numerous doctors and public health officials.  Soon afterwards, it became one of the hardest-hit cities in the US.  (Here is more, from the New York Times.)

In a sense, we are both better and worse off than those who experienced the Spanish flu.  On one hand, our medical science is more advanced; we can now produce vaccines against new influenza strains, albeit at a delay of several months.  (Because the flu virus mutates rapidly, older vaccines, including the one prescribed this past winter, are ineffective against emergent strains such as this one.)  We have also learned the importance of quick and sustained public health measures: witness the recent warning against traveling to Mexico as a result of the disease.

However, modern transportation makes it easier for civilians to pass the flu from country to country, making it harder to isolate the disease to a single region.  In 1918 the flow of soldiers throughout the US and between the US and Europe are credited with helping to spreada disease that originated in the Midwest to every corner of the globe.  This time it may be passenger jets, not steamships, that spread this emerging pandemic.

New swine flu outbreak

For years we’ve been worried about a bird influenza strain (H5N1) mutating to infect humans and permit human-to-human transmission.  Now it appears that a pig flu (H1N1) has in fact adapted to humans, infecting almost a thousand in Mexico and several in the US, and causing nearly 100 deaths.

The United States government declared a public health emergency Sunday as the number of identified cases of swine flu in the nation rose to 20.

The declaration is part of a “standard operating procedure” that will make available additional government resources to combat the virus, Homeland Security Secretary Janet Napolitano said at the White House.

Additional cases of swine flu are expected to be reported in the coming days, added Dr. Richard Besser, acting director of the Centers for Disease Control and Prevention.

Past flu pandemics have caused up to a million casualties, and the epic 1918 “Spanish” flu pandemic is estimated to have killed up 10% of young adults worldwide.  However, there have been false alarms as well; a few cases in 1976 prompted massive immunizations in the US for an epidemic that never materialied, and the vaccine caused a number of adverse reactions.

Unlike earlier epidemics, however, it appears that public health services are being more proactive, both in monitoring the disease and in responding, in Mexico, with closure of gathering places such as schools.  It’s thus very possible that we will avoid repeating the mistakes of past pandemics that led to mass casualties.  However, this is a story to keep an eye on.

Here’s a presentation that two classmates and I did for an infectious disease class.  It talks about the experience of the 1918 flu pandemic, the biology of the influenza virus, and the precautions necessary to prevent a new pandemic (bird flu in this case, but relevant to today).

Masks in Mexico City subway

Learning: so easy a slime mold can do it!

This is pretty amazing.  Single-celled slime molds have been shown to learn to anticipate events after repeated stimuli.  In this case:

As the cells crawled across an agar plate, the researchers subjected them to cold, dry conditions for the first 10 minutes of every hour. During these cool spells, the cells slowed down their motion. After three cold snaps the scientists stopped changing the temperature and humidity and watched to see whether the amoebas had learned the pattern. Sure enough, many of the cells throttled back right on the hour in anticipation of another bout of cold weather.

This sort of learning has been demonstrated before, in organisms as simple as earthworms and planarians.  But even then, these animals have at least a rudimentary “brain” consisting of specialized neurons.  Slime molds, on the other hand, are independent single-celled organisms.  They do, however, have the ability to aggregate to form complex systems, migrating as a single unit and forming a stalk to sporulate.  This new finding suggests that these emergent properties can also produce intelligent behavior, without the need for a nervous system.

For a bonus: detailed and colorful pictures of slime molds, with a focus on their fruiting bodies.

Smarter than most lower eukaryotes, including Congressmen.

Computer discovers the laws of physics

In just over a day, a computer program at Cornell has extrapolated Newton’s laws of motions from a pendulum’s swings. The program starts with random combinations of addition, subtraction, multiplication, division and a few algebraic operators. Then, via a genetic algorithm, the program refines the formulae to pick out ones that fit the data better. It came up with the law of conservation of momentum and Newton’s second law of motion.

This raises some serious questions about the nature of science. It took physicists centuries to develop laws of nature to fit observed data; it took Hod Lipson’s program a day. We could worry about machines replacing human ingenuity. Or — a view I find more compelling — we see these algorithms as an aid to deriving the laws that explain other agglomerations of data, such as the genome or proteome. We don’t know the “laws of biology” in the way we know the laws of physics; and still less do we know the “laws of social science.” Maybe artificial intelligence is the best way to make progress on that.

Here’s Lipson giving a TED talk on robotics.

“What we have [to avoid], is a failure to communicate”

From SEED magazine:

Recently, a small group of American and Chinese scientists and engineers collaborated on a compendium of roughly a thousand terms and phrases related to nonproliferation, testing, and more. The latest edition of this “Nuclear Security Glossary” was made freely available online in November, though it remains a work in progress.

The need for such a nuclear glossary — a joint effort of the US Committee on International Security and Arms Control (CISAC) and the Chinese Scientists Group on Arms Control (CSGAC) — arose because accurate translations between English and Chinese can be tricky under the best of circumstances, and in the highly technical context of nuclear terminology, they are of fundamental importance. “Science rests as much on communication as discovery,” says Raymond Jeanloz, a physicist and CISAC member who helped craft the glossary.

While science is often an international collaboration, scientists remain the products of single nations and cultures.  They must therefore make a special effort to communicate with each other, particularly as science involves so many neologisms that complicate translations.  Many proteins, for example, have multiple names as they are first discovered in high-throughput assays (with little known about them) and then gradually characterized so they can be named based on function.

As a result, simply keeping up with and systematizing this torrent of new information is a task in itself – particularly, as the article notes, when there is a language barrier on top of it, and the science has political implications.