Tuesday, February 25, 2014

An Academic Writers' Block

I'm sitting at my desk as I would in a trench. I am wound tight, my breathing is irregular and highly suspect when conditioned on the lethargic style of my workday. I don't actively seek out tasks to procrastinate, but do actively take them up when they present themselves to me (and they do, at a relentless rate). For the first time in my academic career, I think I am experience writers' block.

I can't speak for anyone else but myself: I am in uncharted waters. As a graduate student, I enjoyed writing papers, and because I started out doing arduous mathematical calculations, I frequently texed up my results immediately. This rapid writing had the double benefit of getting stuff written faster, and allowing me to check my calculations immediately for errors. This meant that I was very quickly able to produce drafts of publications which were eventually published (years later) to critical acclaim among the incredibly large community of 5 scientists, 2 parents, and countless (8) labmates who may have read them. I continued writing, and always had a paper in review, without interruption, from my second year of graduate school until a few weeks ago, when the last one (for which I was a quaternary or so author, and contributed little) was accepted.

Given how much accomplished in those 5 years of graduate school (a whole 30 or so citations!) I don't know how I got here. I have results, I think they are reasonable for publication, and I can't bring myself to actually compose a manuscript. There is no reason to wait: I believe that the results are correct, and for some population of scientists perhaps even interesting. Some of the analysis is novel, and it's scale is quite large given prior work. In fact, I have two separate results, two separate papers. So why the hell can't I write?

It's a perplexing situation I find myself in. I have been reading unconventionally uncontemporary literature recently (the youngest piece is Irving's A Prayer for Owen Meany, and I've also ripped into Sinclair Lewis and Steinbeck). The prose of these three authors is incredible, it sometimes brings me to my knees. In the past, reading literature of this quality while I write has inspired me, has led to steal stylistic elements of language for myself. But now, I feel utterly drained. I can't find the words.

So I have come here to this blog to get myself started. To write a post about how I can't bring myself to write. Let's see if it works.

Tuesday, December 17, 2013

How We and They Should Really Feel about Evolution, Part 2

In my last post, I wrote about a technique, employed by a few different groups around the world, that infers protein structure by measuring how pairs of amino acids co-evolve across different organisms. In my opinion, this story frames (so perfectly) one of the tenuous issues with creationism. Ultimately, it boils down to semantics.

I hate semantics. When confusion results from simply not explaining yourself well, I start to yawn. I care about the truth, and I'm usually disappointed when the truth is obscured because of unclear or misleading language. This is, at least in my opinion, one of those cases. Let me elaborate:

Creationists, in many cases, do not believe in the existence of evolution. As with all generalizations, it is not true that all creationists deny the existence of evolution; some refuse to believe that complex structures could have evolved, while admitting that evolution in the lab exists. I can't say with any certainty what proportion the community believes that, and I have little to say about it. Instead, I want to address those who simply don't believe in evolution at all, including that which takes place in a lab.

It is, simply, scientific fact that evolution takes place in the lab. It occurs on a time-scale along which humans can actually observe it, and it's results are quantifiable. Hypotheses predicated on the existence of an evolutionary process are proven correct, and experiments of this sort are so common that one cannot count the number of times they've been repeated. So why all the hubbub? If this aspect of evolution is so clearly true (and well-accepted by some proportion of the Creationist community), lets all agree that, at minimum, this minimalistic recipe for evolution should be taught in schools.

Why does it matter whether we teach a little, if most of the meat remains censored? Because human curiosity, and especially the curiosity of children, is hard to tame. Teach a child the basics of genetics, inheritance, and the rules of evolutionary selection, and their mind will eventually wander to the same basic questions being tackled in active research today. Tell a child that we can evolve a bacterium to grow faster, and they will wonder if we can evolve a human to, as well. Plant the seed, and let it grow.

Wednesday, September 25, 2013

How We and They Should Really Feel about Evolution, Part 1

I'm just returning from a conference about protein structure prediction. The common thread uniting the research of the attendees was this: that, by using evolutionary information, one may be able to fold proteins. A little more detail may be needed: protein structure prediction (predicting the 3D shape of a protein) is a difficult problem, perhaps the grand, unsolved problem of computational biology. As it turns out, one way to fold proteins (as proposed by those at this conference) is to look at the common features of the same protein across many organisms.

Proteins are long strings of molecules called amino acids, and one might expect that if you see the same amino acid in the same spot in every species, then that's probably a pretty important amino acid. This intuition is accurate, but what if we take it a little further? Suppose that we observe that a pair of amino acids, quite distantly separated on the same protein, seem to exhibit a common pattern when we look across many organisms. An example: if the amino acid in position 1 is X, then the amino acid in position 2 is always Y. If the amino acid in position 1 is A, then the amino acid in position 2 is always B. So, while for a single position, one may see lots of variation in the pattern of amino acids, a clear pattern emerges when looking at the pair.

What does this mean? That's the interesting part. Why might two amino acids that are far apart ever appear to behave according to a common pattern? The answer is that while a protein is indeed a linear chain of amino acids, this chain folds up in a complicated way. Indeed, when this chain folds, those amino acids may in fact be physically touching each other! Then, the choice of amino acid in position 1 may have a huge effect on the choice of amino acid in position 2, because these amino acids will be physically interacting with each other.

OK, how does this have to do with protein folding, and ultimately with evolution? Well, one can use this type of information, these patterns of "co-evolution" (to be explained in a minute) to identify which amino acids in the chain are in contact with each other. This type of information, known as a "contact map," can rapidly accelerate how quickly we can fold the protein, because now we know (to some extent) which amino acids are next to which other amino acids.

So what does this have to do with evolution? Well, like I said, these patterns we observe are identified by looking at the amino acid sequence of the same protein across many organisms. How did this protein manage to end up in all of these organisms? If we look back, far into the history of life, these organisms likely shared a common ancestor, which harbored an ancestral version of this protein. In the course of evolution, distinct species arose from this ancestor, and the sequence of the protein evolved differently in each of these species. Comparing the course of this evolution gives us precisely the information I just described, and potentially allows us to fold proteins.

In the next post, I'll try to describe why this little story frames very well the disconnect between the creationists (and their opinions on what we should teach in schools) and the questions many biologists ask today regarding evolution.

Sunday, September 22, 2013

On Giving Up

It's difficult to admit defeat. Surprisingly, at least in science, it's harder to know if you are truly defeated. The scientific method, or whatever bastardize version of it is implemented by the droves of present-day benchtop and desktop scientists, depends on the proposal and evaluation of competing hypotheses. In the course of the scientific method, many (indeed, most) hypotheses should be defeated. When you are feeling around in a dark room for the light switch to illuminate the truth, you'll more often that not be tripping over the rug. The strange thing is that, in my experience, I frequently ask questions for which I really, really expect an affirmative outcome. I can't say that I don't fail, but I rarely fail big. 

So it's interesting to write a blog about giving up on a project. For the last three months, I've spent a great deal of time asking a very specific question about cancer metabolism. The upside of asking this question was that I knew nothing about how to answer it. I had to learn about chemotherapy, about generalized linear models, about coding in Python. I had to learn how to read cancer literature, and about the caveats of the cancer literature. And as I asked this question, I had a repeated series of ups and downs, of very small p-values and very large p-values. I learned about the details of my data, and adjusted my models, frequently shifting from one modeling framework to another. In short, I was a cancer ninja.

Then, after describing my results to a lab mate, I realized that they were, in the fine words of Mr. Kurt Vonnegut, doodley-squat. Don't get me wrong, the results were as good as they were going to get. I know the ins and outs of this thing, and my models are good. The problem was that, for the question I was asking, the data wasn't good enough. This became clear to me in an interaction which is probably quite familiar to many computational scientists working alongside experimentalists: we make a prediction, think its great, and show it to someone who we want to do an experiment to validate our predictions. The experimentalist takes one look, and says…"That's it? Really?" Maybe in not-so-few words, but you know the tone of voice.


Now comes the inevitable disappointment. Why did I spend so much time and effort working on a project doomed to fail? Why didn't I see this coming before? Eventually: how can I learn front his so I never fall into this trap again? What I've come out of this experience with is a bit of old-timer, veteran knowledge: sometimes you just can't know something is doomed until you try. It seems obvious, but it is hard to swallow when you are the one with your lab notebook covered in doodley-squat. So it goes...

Friday, July 12, 2013

The PostDoc Sweet Spot

I recently watched a great talk by Uri Alon at TedX, where he described his belief in "the cloud." The cloud is that moment during your research where, after many failed experiments, you feel frustrated and lost. Uri (and what a name that is, sounds kind of Zen-master like) instructs us to take a breath, acknowledge we are in the cloud, and embrace our misery. Being in the cloud means you are literally sitting at the fringes of human knowledge, and whatever stands beyond the edge is unknown and beautiful.

After hearing the talk, I naturally sought out more of Uri's advice. I found an essay he wrote a few years ago about choosing a good project. Besides the cloud, he discusses the Pareto frontier of research projects. Uri's coordinate system has two axes: simplicity and impact. Depending on the point you are at in your career, you should choose a research project that lies in different areas of this pseudo-Descartsian plane. For beginning Ph.D. students, choose an easy project, he writes, one that will build confidence, even if the impact may be small. For PIs, shoot for big impact, even if it will take time to succeed.

But for postdocs, the advice is a bit different. He recommends thinking long and hard (his rule, not even a rule of thumb) is three months. Plan well, and choose a problem which will have big impact but can be solved quickly. This is where I am right now: thinking. Thinking thinking thinking.

Wednesday, July 10, 2013

The Story with Gliomas and Backtracking Literature

A few days ago, I attended a great talk here at MSKCC by Timothy Chan on gliomas. Gliomas are tumors of the brain or spine that arise from glial cells, and (according to Wikipedia) make up about 30% of CNS tumors.

OK, so why was the talk so interesting? Well, I didn't know anything about gliomas before, and now I do. One of the things I learned is that many gliomas seem to have the same pair of enzymes (IDH1/IDH2) mutated. It is well-known now (but not a few years ago) that one consequence of these mutations is that the mutant IDH's produce 2-hydroxyglutarate (2-HG), instead of alpha-ketoglutarate. OK, gibberish gibberish, what does it mean? Well, 2-HG does a lot of bad stuff in the cell, and seems to be implicated in tumor progression. Most notably (to me), it seems to activate/inhibit a bunch of enzymes involved in methylation and demethylation, which eventually cause what Dr. Chan calls "epigenetic chaos." The epigenome of these gliomas just looks totally screwed up in comparison to the epigenome of normal, non-cancerous cells.

Again, so what? Well, the epigenome is what you might imagine as a very high-level form of regulation. Epigenetic marks on DNA can silence or upregulate the expression of genes. Aside from the really fascinating link between metabolism and epigenetics here, what really dazzled me about Chan's talk was that it led me to re-evaluate everything I have been reading as a new postdoc. I have a paper tacked to the wall of my cube titled "Cancer mistunes methylation," and while it is heavily annotated and marked up with my notes, I'm surprised to find that after Chan's talk, I need to go back and read again. And that is the mark of a great talk; I realize that everything I thought I understood, I really didn't. And now here I go down the rabbit-hole.

Monday, June 17, 2013

Steinbeck and Science

"Communication must destroy localness, by a slow, inevitable process. I can remember a time when I could almost pinpoint a man's place of origin by his speech. That is growing more difficult now and will in the foreseeable future become impossible...Just as our bread, mixed and baked, packaged and sold without benefit of accident or human frailty, is uniformly good and uniformly tasteless, so will our speech become one speech." - John Steinbeck, Travels with Charley

Steinbeck could write. Although he was referring to the extinction of our various American flavors of English, I found his words oddly inspiring as I pondered modern science (what a weighty topic, "modern science"). The process by which ideas gain popularity and acceptance is not necessarily through healthy scientific discourse. I frequently wonder of how many good ideas are thought up by those without sufficient acclaim or visibility to enable those ideas to gain traction. The way the system is rigged, ideas which are published in high-visibility journals are those which are discussed, but how else could it be?

What reassures me about this is that, as I travel to conferences and talk to other scientists in the field, I get the distinct impression that their training had its own dialect, its own flavor, distinct from my own but with shared ingredients. While we are all systems biologists, the anecdotal classroom discussions we cite are different. It used to be that many people I met quoted Uri Alon's book, but now, the literature is so vast and the resources so plentiful, that the book seems outdated. And that is nice; many perspectives, all poised on the same problem, may yield many distinct recipes for solutions.