New research model: the “Flash Mob” style project

One of the bottlenecks to excellent quantitative work is the looooong arc of the publication process. While this is obviously a necessary evil of being a productive scientist (and perhaps “evil” is the wrong word; peer review can be very enjoyable), it would be nice if there were other research models.

Thankfully, with the rise of preprint servers for manuscripts, other models of research are being explored and engaged. Recently, postdoctoral fellows from the Santa Fe Institute participated in a ‘72 Hours of Science’ Project where all aspects of the paper–the idea, the method development, model construction, analysis, writing, and publishing–were accomplished in a 72 hour window, leading to a super cool paper about “beneficial epidemics.”

We might talk more about the paper in the near future, but we mostly wanted to highlight this style of research. I’ve heard it referred to as “Flash Mob” style research (I first heard it used by Peter Dodds, formerly of MIT, now at the University of Vermont/Vermont Complex Systems Center).

Here at the Open Modeling Group would love to engage this model for developing some models….stay tuned!

 

New in ‘Nature Physics’: A surprise driver of epidemics!

Some great new work was recently published in the journal Nature Physics by a colleague, Professor Sam Scarpino of the University of Vermont and the Vermont Complex Systems Center. In it, Scarpino and others reveal a hidden, classically overlooked driver of epidemics: the replacement of sick people with healthy people at key points in an infection network!

This work is the rare study that is both highly surprising and completely intuitive: it’s surprising because it’s….weird to think that healthy people at certain points in an infection network drive epidemics…think about it: when your school teacher falls ill, and is replaced by a healthy substitute…that…increases the number of people who become infected? Really?

It is intuitive, however, because healthy people are more likely to contact more people before they are overcome with clinical symptoms.

It has many implications, the first of which involves the prediction and forecasting of epidemics.  There are other implications for the public health realm, for example, if the replacement of certain individuals in settings is prolonging or amplifying epidemics, maybe we need a more informed process for replacement (“relational exchange” as Scarpino and company call it).

We (the Open Modeling Research Group) might dive further into this soon, but we wanted to highlight this great new work, and point you towards the actual manuscript and reports in several news outlets (like here, and here).

Update: an even OFFL-er version, in PLoS ONE!

If our ArXiv manuscript wasn’t bad enough, apparently some people with really great jobs thought the manuscript deserved to be published in a real, grownup journal.  Yeah, we can’t believe it either.

It can be found here:

We’re excited to see our work in print, but are more eager to get going on some applications of the OFFL model. We’re kicking around a few ideas these days: plant invasion, Hepatitis C virus and…The Avengers? Any other weird things you want to model, think we can model, want to help us model?

The most OFFL paper we’ve ever written

It’s done! Brandon and I have finally posted our first paper as a group at the Quantitative Biology preprint archive:

OFFl models: novel schema for dynamical modeling of biological systems

We introduce our new formalism for helping bio/medical practitioners and students (and anyone else) organize their observations of the world and collate them together into a mathematical model, all without them having to deal with any actual math. It uses a formal grammar of pictorial flow diagrams to automatically generate a system of non-linear ordinary differential equations. What to call this schema of “ODEs and formal flow diagrams”? It’s OFFl, of course! And OFFl is the new awesome!

We’ll have a lot more to say about OFFl in good time, but for now, we’re glad to finally have it out in the wild. We’ll update this space when the journal version finally gets published.

And, yes, we plan to apply OFFl to understanding the zombie apocalypse and any other existential threat or eschatological crisis the world needs scientists like us to save it from.

Back off, man ... I'm a scientist.
Back off, man … we’re *all* scientists.

Huzzah!

UPDATE (2016-06-07): OFFL now published in PLoS ONE!

I may be a barbarian at the gate, but I swear I won’t burn down your city, and I won’t come in uninvited.

Thanks to a Facebook post by Physics Today magazine, I recently read this very topical essay by over at undark.org.

Physicists at the Gate: Collaboration and Tribalism in Science

It’s pretty great. Go read it. OK, done?

It takes direct aim at people like me: physicists moving into complex established fields and feeling like we can do something useful because we know more math. The author’s use of “tribalism” in the title is a bit off in my opinion, but otherwise I find the discussion of the issue to be both accurate and comprehensive enough to qualify as standard reading for any physicists looking to branch out like this.

As I’ve started getting my feet wet in systems biology, I’ve tried to be particularly sensitive to the dynamic discussed in the essay, especially in understanding that before I’m even introduced as a physicist, I’m preceded in reputation by all of my colleagues whose naive arrogance has helped generate this stereotype of our field in the first place. For my own part, I just make a point of being respectful and humble. Most of importantly, I listen. I listen a lot. After all, even if I can’t yet understand what makes the biologists so excited about the questions they are asking, I can’t help but yield to the empirical fact that the global program of biologists pursuing basic curiosity-driven research according to what excites them and not me has been amazingly productive by whatever measure of productivity you care to use. It’s enough to keep anyone humble, even a physicist. So, I figure, if I listen, I might learn something.

That said, returning to the essay, it’s actually a bit of a myth that physicists only study simple things and that the success of the field is largely due to concentrating on problems that are easy enough to be approachable by mathematics. This may describe theoretical particle physicists, sure, but I can assure you that most working physicists are studying systems which are so complex that there is nearly no hope of ever understanding them from a reductionist, first principles description of the system’s microscopic constituents. (In fact, not even theoretical particle physics is like that, but that’s another essay!) In fact, our human experience with the everyday world of bulk matter is dominated by behaviors which are not even present in the basic constituents at all, but are somehow abstractly encoded in their arrangement. (They are “emergent”.) Physicists study the heck out these things because, just like every other scientist, we are motived in our work by studying the hardest questions we think we can make progress on. One doesn’t need to go to all the way to biology, ecology, and archeology to find problems which are historically contingent, intricately detailed, or otherwise just too hard to model as a spherical cow on a frictionless plane. We don’t even know how friction works.

Nevertheless, it’s true that part of the intellectual training and common culture of physicists is to look carefully at a system and then, once you think you’ve started to make some sense of it, start stripping away the unnecessary details until you’re left with the essential core of the problem. Then you start going at it with your great big toolbox of problem-solving tricks, some of which involve a fair bit of math. In general, this has been a massively successful program and is a big part of what makes physicists so employable outside of academic research.

The problem with this approach is that the notion of which details are essential and which are unnecessary depends in a fundamental way on the questions you are asking about the system. The questions you ask depend as much on you and your prior intellectual biases — your paradigm, in Kuhn’s sense — as they do on the system. And, if you’re trained as a physicist and looking at a new system outside your past experience, you’re likely to decide that the unnecessary details are the ones which you don’t yet know how to ask questions about. You’re likely to decide that the interesting questions are the ones amenable to your toolbox. You may even dismiss whole lines of questioning that form the basis of entire research fields as “uninteresting details”. You are unlikely to convince the people already working in those fields to drop their extant successes and take up your glorious new insights.

In other words, your hard work will have no impact, another wasted effort on the scrap heap of scientific history. I don’t know about you, but I’m not in science for the fame and fortune. I’m here because I want all those irretrievable hours of my life that I’ve spent on work to amount to something. I want my efforts to have positive impact on the human condition. So, to avoid the honest-to-goodness tragedy of science without impact, I’ve been trying my best to listen to what people who already know what they are doing think is important, and then searching hard for the words to put my own work in their context. I don’t just want to start doing science the way a biologist would, of course, because that would defeat the point of being an interdisciplinary voice. But, it seems important to express my outsider voice using insider words in order to be heard. At least, that’s the plan for now. We’ll so how well this turns out in time.

If nothing else, all this listening is doing great things for expanding my world view, and that’s opened up whole new avenues for thinking about my physics problems. So, there’s another advantage of not just crashing through gates like a naive barbarian and demanding to be heard: it actually makes me a better physicist. Who knew?

Zombies

Yup, zombies. Here at the Open Models blog, we’re big fans our random walking friends from beyond the grave. They’re a never ending source of inspiration for different twists and turns in epidemiological, ecological, biophysical, and other kinds of model building.

And, we’re not the only ones that think so! There’s a growing technical literature on serious zombie studies. To show our own devotion to this critical and timely research field, we’re going to start maintaining a zombie studies resource page here on the the blog.

Enjoy! Tell your friends! Let us know if you find it useful, or if there’s important zombie research we’re missing from the list.

Zombie Studies Resource Page