The Futile Cycle

I was sitting in my lab reading through the PNAS early online publication articles when I come across a letter to the editor with the title: “Going beyond the genetic view of cancer.” (PDF) Intrigued, I read it, and to my horror, I find a letter filled with words like this:

Dynamic protein-based phenomena–for instance, (insulin-driven) “oncoprotein metastasis” explicable by an extension of physical string theory into (sub)cellular biology…

The fact that the author (a one “Razvan Tudor Radulescu”) cites an arXiv “pre-print” that he wrote himself is quite suspicious. His contact information, of course, is with non-institutional, free email address.

To me, this can mean only: HE’S A CRANK.

What is PNAS doing publishing letters from cranks? I get spam in my email from this gentleman in Nicaragua, should I publish those in a journal, too?

Perhaps I’m leaping to conclusions. Let’s take a look at the arXiv pre-print, which would be more fleshed out than a single letter to the editor:

Here, a new scenario is put forward on the spreading of the neoplastic process across cells and tissues that may prove seminal both for our future understanding and treatment of malignancies.

First sign of crank-dom: exceedingly high opinion of one’s own work, describing it as “seminal.”

In this context, my peptide string theory (10-12) is likely to represent a significant addition [to the field]. It rests upon the assumption according to which major biological processes concerning distinct, yet related proteins are the result both of (long-distance) attractive forces in the sense of the physical string theory (13,14) and of “emergent properties” inherent to the same proteins whereby the term “emergent” is to be understood as employed by John Searle in his book entitled “The Mystery of Consciousness.”

The sentence continues, but the point is rather obvious. Crank. What are the citations to? Let’s see…citations 10-12 are self-citations (one to yet another arXiv preprint). Citation 13 is to a popular science book, The Elegant Universe, and citation 14 is to an editorial by Ed Witten.

Now, I’m no expert in string theory, but I doubt that these second two works have much biological insight to offer. String theory, after all, is about understanding how all the various laws of physics might be unified. Sorry, but general relativity? Not all that relevant to what goes on in the body, since biology doesn’t really go near the speed of light. Quantum mechanics is more related, in that it’s the foundations of chemistry and of molecular-level phenomena, but even then, we’re not pushing the boundaries of our knowledge of sub-atomic processes in looking at biology.

Why is this letter being published by PNAS? All I know is, this kind of stuff makes real scientists look bad. Does the National Academy not vet letters to the editor before publishing them?

Pure Pedantry links to an awesome paper on visualizing the cell cycle in mice using fluorescent markers. They can even look at the cell cycle state of cells in histological sectionsof mice! Check out the blog post, because Jake Young highlights the coolest picture and the coolest video in the paper; the movie at Cell and ScienceDirect seem to be down, so you can see the video at Pure Pedantry or at Google Video.

I have yet to take my general exam (a.k.a. qualifying exam), which is what I need to pass in order to get to the next stage of my Ph.D. studies. Up until now, I’m basically a masters’ student, doing some research, taking classes, attending seminars, and basically learning my way around being a research scientist. After generals (assuming I pass…), my department expects us to teach and focus on completing my thesis.

I know some schools don’t do this, but in my department, the general exam is basically an oral exam — we present our thesis proposal, as well as one paper that is on a related (but different) field. The prospect of it frightens me to no end, but it’s still about a year away.

Of course, the main thing that I’ll be judged on is my science, but humans are humans, and since the exam is quite subjective, I’ve got to use every edge I’ve got, including improving my speaking skills.

At MIT, there’s a famous annual lecture given by Prof. Patrick Winston called “How to Speak.” This year, a blogger managed to take some notes from the lecture, for those less fortunate of us who don’t have a chance to see it. There are even some tips on speaking specifically for oral exams. I like this one:

3. Practice. Ask your friends to listen to your talk. Tell them to try to make you cry.

Today is apparently Blogroll Amnesty Day, in which small blogs link to smaller blogs, and so on. Alas, my blog is one of the smaller ones around, but since I tend to read across genres pretty broadly, perhaps I can help some people find some new blogs that they’d not seen before.

New to the blogroll, which has been un-changed for a very long time:

Computational Biology News

This blog is mostly links to interesting articles rather than a bunch of articles themselves, but Animesh Sharma nicely curates and finds stories on the net that I wouldn’t find otherwise.

DrugMonkey

DrugMonkey, which was recently assimilated into the SciBorg collective, is a great blog on health regulation in America.

Ars Technica: Nobel Intent

I’ve read Ars Technica for almost as long as it’s been around, starting from way back in my days in which I was really into computer science. Basically, I thought of Ars Technica as the thinking man’s Slashdot: less frenetic, more in-depth, and not pandering to the masses. Lately, Ars has gotten more pop-news-ish, but for the most part, it still seems to retain its journalistic integrity. Nobel Intent is Ars’ science journal, and they’ve got some great writers there. Apparently, a lot of science blogs have never heard of Ars Technica, so check it out if you haven’t read it before.

UPDATE:Forgot to add Evolving Thoughts, which covers biology, the philosophy thereof, history, politics, creationism, and really a whole swath of things that I find interesting.

In the first major casualty of the etBLAST algorithm and Deja Vu database has been found at Harvard Medical School, where Prof. Lee Simon’s review paper has been found to have large sections copied from another professor’s paper.

I had hoped that Deja Vu would consist of articles from random foreign countries and small, obscure universities, but alas, I was perhaps a little naïve. Perhaps the good part about this will be that it encourages authors to be much more reluctant to plagiarize.

On the other hand, it depends on whether the journals care. Elsevier, in this case, did the right thing and acted upon the evidence to retract the paper, but other journals don’t have such “enlightened” policies. I heard once about a professor that was reviewing a manuscript for a journal when he found that the other author had plagiarized sections from one of the professor’s own papers! When the professor notified the journal editor, they informed him that this was commonplace, and that he should just review the article anyway.

What floats to the top in science is often beautiful, but there’s a lot of crap that sinks to the bottom.

Via Ars Technica, there’s apparently a new database online called Deja Vu, which catalogues duplicate (or near duplicate) entries in MEDLINE, using an algorithm very similar to BLAST.

The majority of the entries retained in the database seem to be cases of outright plagiarism. Going through a few of the entries is like dredging the bottoms of the academic pool; most of the journals are pretty obscure or quite poor in quality, and I bet most of the authors are pretty unknown, too. It’s a strange world down in the pits…

Alas, I (as a poor, time-crunched graduate student) will not be able to make it to the North Carolina Science Blogging Conference, but they have a nice website that’s going to be aggregating a lot of the content surrounding the conference.

(Beware of that aggregating website; it seems to break on Safari, so use Firefox. Alas, Mac users do not seem to command respect and deference quite yet.)

Over at CoyoteBlog, I’ve made a few comments on this blog post in response to a quip that the author was skeptical about global warming. I’d like to repost my arguments here, since I think it’s a general reflection of my thoughts on science:

I think that trusting scientists on their views in their fields of expertise is generally a good idea, and if the IPCC says that global warming is starting to accelerate, I’m inclined to accept it; it’s really the only intelligent thing to do. Policy and normative notions, one can debate, even if you aren’t an expert, but for facts and other positive notions, I’ll stick to the experts, and they seem to say that global warming is real.

If everyone trusted the experts more, the world would be a better place. None of the pseudo-science homeopathy/complementary medicine nonsense, none of the liberalized ignorance of basic economic facts, and none of the nonsense of evolution denial.

Here, someone replied to me:

Your response would make sense if you didn’t take into account the idea that scientists get way more money to do research about catastrophic global warming versus natural equilibrium.

In addition, I don’t agree with the “trusting experts” argument because I have yet to see an approach to global warming based on the emerging field of complex adaptive systems. As one who studies this field, am I not an “expert”, shouldn’t you trust me? Because I’ve found that complex systems move towards different equilibriums, not tipping points. By classifying the global climate as a complex adaptive system, I have to disagree with the current so-called “experts” who claim there is a tipping point.

Personally, I think this whole argument of “trusting experts” is flawed because one is only an expert until somebody with a new theory comes along that is better. Let’s not forget the “experts” told us that the Earth was at the center of the universe, and they had the majority consensus. It used to be mind boggling to think otherwise.

“Trusting experts” is probably one of those things which will lead to the loss of individual liberty and further reinforces the point of this blog article

Thus, I replied,

Well, notice that I used the plural here. The views of an individual scientist are, for better or worse, his/her own views. The scientific consensus, on the other hand, represents the views of a large swath of scientists, and the competition between scientists for funding and publication generally means that, on the whole, the “most correct” interpretation tends to win. I don’t know how much I can trust an individual scientist’s views, but if a vast majority of scientists support a view, then I’ll stick with that one. I’m not addressing here the whole “tipping point” phrasing or alarmism that the particular journalist used in the above blog post. I’m talking about general skepticism about global warming. Unless I have specific, methodological problems with the general field’s reasoning, and have experience in in that field, I’d say that I will always accept the field’s scientific consensus on a subject.

There’s a word for people who, with absolutely no expertise, think that they have the answers in the face of tons of experts: crackpot. Lay people (i.e. non-scientists) who go against scientific consensus are the same as people who think that they can prove Einstein’s Theory of Relativity wrong, or who think they’ve proved Fermat’s Last Theorem in two short sentences. They’re not brilliant critics; they’re raving madmen. Just because Rob is an expert in complex adaptive systems doesn’t mean he’s an expert in climate science. That’s like saying a mathematician is an expert in cryptography; yes, cryptography uses mathematics, but unless that mathematician has a focus on cryptography, they won’t know all the algorithms, all the details, all the issues involved. If a mathematician were to persist in making sweeping non-consensus claims about cryptography without reading about cryptography, he or she’d be a crackpot, regardless of how smart they are.

So if Rob is going to make some claim about climate science because he’s in a field he thinks is related, he’d better have much better reasons than “adaptive systems converge upon fixed points”, cause climate phase trajectories lie in a large-dimensional manifold, and in such situations, fixed points often abound; in addition, we don’t know the rate of convergence to the various fixed points, and presumably the ever-changing flux of the sun, geothermal activity, and rotation of the earth would prevent such steady-state convergence. Perhaps anthropogenic climate change is driven by the evolution of fixed points in certain directions; perhaps our introduction of greenhouse gases has led to some sort of bifurcation, and that’s leading to the increase in temperatures. Hey look, I can make vague generalities about climate science as a complex system, too! Do you believe me? I would hope you go to the experts instead.

Sure, some “experts” in the past have claimed that the Earth was in the center of the universe, perhaps, but they were not scientists; they didn’t follow the scientific method, they used state-mandated religious views to arrive at their conclusions; religion obviously is not a source of facts, and isn’t a way to find scientific truths.

The IPCC (a coalition of most climate scientists) thinks that currently, mankind’s contribute to the environment has caused a significant change in the climate, leading to a significant change in the world’s temperature compared to the past. The main objections to this that I’ve seen come from 1) people with no expertise, and 2) a few (very few) climate scientists. The people with no expertise include members of the press, politicians, random scientists from other fields, and so on. I don’t think they have valid points to make unless they really read the literature first. That the vast majority of climate scientists agree with the IPCCs statements nullifies my second concern.

And an expert in climate science is still an expert in climate science once new theories come about. Did physicists stop becoming physicists with the quantum mechanics revolution? No, they instead all became expert quantum physicists. Did they leave the theory up to the public to decide? No, they evaluated it themselves, because people intimately familiar with the experimental details and the underlying facts and methodologies are the best ones to evaluate new theories and evidence. Let the scientists in the field determine what theory is sound and what isn’t; I don’t think the uninformed public should.

The simplest answer isn’t always right, but it’s often one of the most convincing answers. John Timmer over at Ars Technica talks about the rise of a very simple answer to a whole swath of questions that include

1. How do transcription factors, splicing factors, and other such DNA and RNA interacting proteins operate accurately when they’re so promiscuous in what they choose to bind to?
2. Why is so much of the DNA transcribed to RNA?

The simple answer that John gives is simply, transcription and splicing factors don’t operate that accurately, and most of the DNA is transcribed to RNA because of this inaccuracy. Simple! And quite a compelling theory, actually, something I like very much.

The question this line of reasoning raises, then, is whether the RNA editing and selective degradation pathways are sufficient to prevent something bad from happening to the cell. There’s nonsense-mediated decay, signal-sequence localization, non-stop mediated decay, and complicated RNA localization and export control. That’s in addition to RNAi, the evolution of which may have been driven in large part in order to weed out bad RNA.

This would, perhaps, explain why RNA editing features are so much more complicated in mammalian cells than in Saccharomyces cerevisiae, as S. cerevisiae (baker’s yeast) tends to have very highly conserved, high-consensus splice sequences (i.e. low error, but also low complexity), as well as more well-defined and static chromatin and no RNAi.

The story vindicates that complicated gene regulation and patches and makeshift fixes for errors are two sides of the same coin.

There’s a neat blog experiment here for blogs to vow to only post about science, at least one post each day, between February 4th and 8th. I’ve signed up. Will you?