Thursday, February 25, 2010

Where do you learn?

Since gaining actual friends at college (I know, I know), I've been repeatedly surprised by how often people skip lecture. I must have noticed this earlier -- after all, I did observe that lecture halls tended to get less full as the semester went on -- but somehow I thought that all good students always went to lecture or some such. Apparently this is not the case.

Also, it seems like people are supposed to know by this time whether they're "visual learners" or "audio learners" or some such. I have never been able to figure this out. But apparently I learn pretty well by going to lectures (as opposed to doing the reading), although this is probably a motivation problem rather than anything else.

For 2.5 years this strategy served me well, but now that I run into courses with mediocre-to-poor lecturers who are always unclear and often omit large quantities of important material, it's becoming problematic.

Thursday, February 18, 2010

Learning to read all over again: Addendum

The project of learning to critically read the primary literature is going pretty well so far. I'd like to add another stage to the four stages I mentioned:

5. Selectivity. At this point you become able to read "the literature" as a whole, not just individual papers. You can make a good guess about what's important without having to read everything in excruciating detail.

I'll also note that the stages overlap quite a bit. I'm still mostly in Credulity, but there are some papers I can actually criticize, and some I still fail to understand.

Wednesday, February 17, 2010


The water dispenser outside my lab has a cold spigot and a hot spigot. The hot spigot dispenses water which is actually quite hot. However, I couldn't figure out how to use it at first... until I realized that you have to put the handle in a certain position, which is not its natural low-energy gravity-ordained position, and then press it. It's a safety interlock! Good thinking! *makes tea*

I drove in snow for the first time recently. Well, not snow, but there was some kind of sleet-like substance on the roads and falling (slowly) from the sky. It was also the first time I ever skidded (a small amount). I've been on a skid mat, but it was a while ago and I remember it very poorly. I think the problem with the skit mat was either that I was unable to actually skid, or that it made no sense out of the context of a street and the slight hazard of hitting parked cars. There's no spatial reference on an infinite frictionless plane, after all.

Plates of fermenting bacteria on a phenol-red containing medium have this wonderful tendency to grow into miniature sunrises (although they do develop a rather off-putting smell).

Figure 1: Like viewing the sunrise from a mountaintop outhouse.

I just reread Terry Pratchett's Going Postal, and I really liked it. He gives the sort of lovely far-off impressions of the world of the clacks towers that I want to go there but that might destroy the mystique -- you know, the same thing people always say about why the Silmarillion is written in such a dreamy mythic tone. And if Thief of Time is the Discworld's The Matrix, then Going Postal is its Jargon File.

Tuesday, February 16, 2010


I was reading my perfectly innocent-looking homework assignment, when all of a sudden I saw this:

Figure 1: Augh!

You know, I figured a biomechanics class (which is, after all, all about the realistic physicality of biology as opposed to magic cartoon enzymes that always work) would feature more realistic estimates of a bacterium's size.

Figure 2: I also figured it would feature less fleeing and primal terror. [Source]

Of course, the answer is that E. coli are measured in μm (micrometers), not meters, and that the micro sign simply failed to render due to some strange failure of MS Word. The lesson is clear: Mistakes in character encodings will kill us all. Go forth boldly, my friends, and godspeed.

Monday, February 15, 2010

Journal Club: The enzyme bucket brigade

JE Dueber, GC Wu, GR Malmirchegini, TS Moon, CJ Petzold, AV Ullal, KLJ Prather, & JD Keasling. Synthetic protein scaffolds provide modular control over metabolic flux. Nature Biotechnology 27, 753–759 (1 August 2009) | doi:10.1038/nbt.1557

I read and presented this paper in my lab class this past fall -- and I thought it was just one of the coolest papers I'd ever read. Cool concept, rational design, elegant solution to a Hard Problem, multiple benefits, modularity/composability, real-world results... this paper has it all.

You can look at cells as little factories, taking in raw materials and churning out interesting molecules. A cell's naturally occurring assembly lines are optimized by evolution in various ways, to increase efficiency and decrease interference ("cross-talk") with other processes in the cell. In particular, natural metabolic pathways are regulated so that they don't go wildly out of control and start overproducing whatever chemical, because that would be wasteful and expensive (not to mention potentially harmful).

But when you're putting an artificial assembly line into a cell, you have to undo some of these constraints and not others. You have to maximize efficiency, minimize cross-talk, and avoid making toxic products in the middle of the pathway as much as possible. These goals all line up with the goals of the cell. However, your main goal is different from the cell's goal of "produce just enough": you want to produce as much product as possible. More medicine. More biofuel. More super-protein-material-thing. More whatever. So this should be easy, right? The metabolic pathway is made up of enzymes that convert Chemical A to Chemical B to Chemical C, and you're inserting the genes for those enzymes into a bacterium. Why can't you just put very strong promoters in front of those genes, so you get massive quantities of each enzyme, and massive output?

This simple maxing-out approach causes several problems. First of all, it does nothing about the intermediate chemicals along the pathway -- they could still be toxic, or even just float away and go to waste. Second of all, this approach doesn't bother to optimize the ratio of the two enzymes. (If Enzyme 1 is half as efficient as Enzyme 2, then you ought to have twice as much of Enzyme 1.) Third of all, this doesn't do anything to stop the pathway cross-talking with other pathways. Fourth, and possibly most important, there's no guarantee that forcing each individual cell to make as much product as it possibly, possibly can is the most efficient way to convert cell food into useful chemicals. It's probably more efficient to let the cell divert plenty of energy into maintaining its own health and into spawning more cells, so you end up with more product overall.

To solve this dilemma, Dueber et al borrowed a trick that cells often use to regulate their own pathways. A scaffold is a structural protein that grabs on to all the enzymes in a given pathway, and holds them together into something like an assembly line or a bucket brigade.

A question about scaffolds

There are two bits of signal transduction dogma that have started to bother me. I don't know why I didn't spot this before.

Kinases are proteins that activate other proteins by attaching phosphate groups to them. Kinase cascades (several kinases in a row) are fairly common in all kinds of signaling pathways. Their main benefits are amplifying and diversifying the signal. Since a kinase is after all an enzyme, it can catalyze the same reaction over and over; it can activate many copies of the next kinase in the cascade, each of which can activate many copies of... and so on into exponential growth. That's how you get amplification, turning a tiny-but-important input into a massive cell-wide response. Diversification comes in when a kinase has more than one target. This comes in handy when the cell needs to respond to one signal by doing several different things all at once.

Scaffolds are large structural proteins that grab several other proteins from a signaling (or metabolic) pathway and hold them together. This helps them get their job done more efficiently. They help make pathways specific. If Enzyme 2 is stuck on a scaffold between Enzyme 1 and Enzyme 3, it can't very well run off to some other part of the cell and mess something up.

So, I perceive a slight conflict here. On one hand, it's helpful to diversify a signal; on the other hand, signals ought to be specific. On one hand, enzyme cascades amplify signals by working catalytically instead of stoichiometrically; on the other hand, when kinases are bound to a scaffold, their stoichiometric ratio is locked at one-to-one. What's going on here?

The obvious answer is that each type of signal processing is used where it's appropriate, and all types are appropriate in different contexts. If this is the explanation, then you would never expect to find a kinase cascade associated with a scaffold. But that's exactly what the MAP kinase cascade does! What gives? I thought the whole point of having a kinase cascade was to amplify and diversify the signal, which is exactly what the scaffold seems to be preventing.

Figure 1: MAP kinase cascade shamelessly associating with scaffold protein. Have they no shame? What has the yeast mating pathway courting ritual come to these days? [Source]

Journal Club: The amber-suppressing AND gate

Anderson JC, Voigt CA, Arkin AP. Environmental signal integration by a modular AND gate. Molecular Systems Biology 3:133 (2007) | doi:10.1038/msb4100173

Logic gates (AND, OR, NOT, etc.) are the basis of electronic computation. If we'd like to implement biological computation, one of our first steps has to be implementing similar logic gates using proteins and DNA. That is, we need to make devices that accept a few inputs, perform a logical operation on them, and then spit out the result. In the case of an AND gate, we want the output to be ON whenever both of the inputs are ON, and OFF when either input is OFF. It seems easy, but of course, this turns out to be a lot harder in biology (hence, people writing papers about it).

Figure 1: The result of this paper, if you abstract away all the interesting stuff. [Source]

What makes it hard to make an AND gate out of biochemical parts?

Lack of standard connectors. In electronics, every signal is carried by a current, and every connector is a wire. That isn't the case in biology. Biological signals are typically carried by the presence or absence of some protein that carries out some particular chemical reaction that affects other proteins. This is wildly nonstandardized, and it means if Protein A interacts with Receptor A, you can't just plug in Receptor B and expect things to work.

Fuzzy, non-discrete behavior. It's nearly impossible for a biological system to have a perfect ON or OFF state. Even if a signal is mostly off, there'll be a few molecules of it floating around somewhere. And when you go to turn it on, it'll take time. Basically, biological things tend to vary continuously and not discretely (in large jumps).

Crosstalk. If a biological device relies on some particular molecule, then that molecule is going to be everywhere in the cell. So, you can't put two copies of the same device into a cell and expect them to operate independently. They'll interfere or "crosstalk" with each other in ways you don't expect. In contrast, you can throw down dozens of electronic circuit elements onto a breadboard and they won't interfere with each other because they're separated by physical space. In biology, everything's in the same soup.

Tuesday, February 9, 2010

Learning to read all over again

In my RNAi seminar, one of the instructors (they're both postdocs) noted that reading primary literature is a legitimately difficult thing. He expects us all to progress through several stages:

  1. Incomprehension. At first, it's difficult to even figure out what the authors are talking about, or trace the flow of experimental logic. Sometimes it just doesn't flow quite right; other times you get distracted by all the technical details you don't understand at all or don't understand the need for.

  2. Credulity. Once you learn how a few procedures work and start to understand why people do experiments in the order they do, you can finally pull out a paper's main substantive point from all the noise. At this point you'll believe whatever the paper says, because the graphs look pretty compelling, right?

  3. Savage nitpicking. Next you learn to ask methodological questions: why did the authors omit that control, or choose this particular method of statistical analysis? Why do the experiments in this order and not that order? Suddenly every paper looks like complete crap.

  4. Understanding. You begin to understand what is good and what is bad about a paper, and how it fits into the context of the field. This does require some familiarity with that context, but it's also a matter of general experience.

I think I'm more or less at the Credulity stage. I can figure out what a paper is about, though it takes me a while, and I get super excited about the key results. I've gotten to the point where I can read papers by myself, but I've never seen a paper picked apart, analyzed, and criticized in any detail where I could actually follow the conversation (this is hard in high school). So I think I'll be needing professional help, as it were, to move beyond Credulity.

(I might have gone a little bit overboard with the paper-reading classes for this semester, though. I added it up and I think I'm going to be reading 6-8 papers a week, or about one per day. Wow.)

Monday, February 8, 2010

Courses for this semester

This semester's coursework has two themes: "Let's Read And Discuss Primary Literature", and "Time To Stop Abstracting Away The Physical Nature Of Biology". Both of my required bioengineering core classes are about cells, molecules, and tissues from a mechanical engineer's perspective. All three of my electives are literature-discussion classes on various topics.

  • 20.310 Molecular, Cellular, and Tissue Biomechanics - think introductory mechanics, only all the examples are biological things instead of, you know, steel beams or something. Up till now, I've been accustomed to thinking of DNA as a string of digital (AGCT) information, or maybe as a helical molecule; now it's time to think of DNA as a charged elastic rod, or a randomly-walking polymer. How much force can a motor protein exert to pull a vesicle where it's going? What happens when you push and pull on the cytoskeleton? What's the effect of pressure on cells? How do bones reshape themselves in response to forces?

  • 20.330 Fields, Forces, and Flows in Biological Systems - fluids and E&M, only again all the examples are biological things. How do things like diffusion and electrophoresis work? How can you model the cell membrane as an RC circuit? What's the best shape / flow pattern for this sample chamber so that the most protein binds to the sensor on one side?

  • 20.385 Advanced Topics in Synthetic Biology - this is paired with a freshman design/seminar course, 20.20, which I took two years ago and which rocked my world so hard. It's a great introduction to synthetic biology. The frosh get to do design projects, and because it's surprisingly hard to do this when you've only had introductory biology, the upperclassmen mentor the frosh teams. And when we're not busy mentoring, we have synthetic biology journal club. I'll be presenting a couple of papers and I'm super pumped.

  • 7.25 Biological Regulatory Mechanisms - this just sounds fascinating. All the different ways gene expression or protein action can be controlled. Apart from being cool, this is also highly relevant for synthetic biology. Even apart from that, I'm excited about the lectures. This class involves picking apart the experimental logic of papers in a more rigorous way than I've ever had before, which I'm sure will be good for me as well as being fun. We're focusing on demonstrating results and excluding alternative explanations to an extent that seems to be missing in the modern age of "let's generate a zillion data points and then sift through them". Plus, it's a chance to pick the brains of some aged, sage professors. All in all, it really reminds me of ((my interpretation of) what Raffi said about) learning Talmud.

  • 7.346 RNAi: A Revolution in Biology and Therapeutics - yet another paper reading class. I know nothing at all about RNA interference, but it's extremely important both theoretically and (potentially) medically. I'm also interested in using it to make synthetic-biological parts that don't crosstalk as much as protein-based parts do; we'll see if that's feasible.

I plan to blog a lot more about the papers I read in these classes. It probably won't be a full Journal Club post for every paper because I'll be reading about 6-8 papers a week, but I'll at least try to summarize them. (This is partly for your benefit and partly for mine -- I expect that writing paper summaries for the blog will help me read and understand the papers better.) But I will do full Journal Club posts for the papers that I find most interesting, and definitely for the ones I present. I might also write about interesting things that happen in the other two classes.

Monday, February 1, 2010

Academic identity crisis

My dad and I happened to be talking about construction failures earlier today -- a raised highway section that collapsed in the Loma Prieta earthquake; the tiles that fell off the Big Dig ceiling. He spoke about them in terms of redundancy and single points of failure, using the engineer lingo that he's picked up from reading books about civil engineering. When I'm being cynical about synthetic biology, I think biologists like to use these terms to make themselves sound sophisticated.

It does bother me, though, that I will graduate from this place with the word "Engineering" on my degree but possibly without the ability to analyze a design, find its flaws, and fix them. I will be able to design an experiment to determine whether Protein X affects Process Y in the cells of Species Z, which covers the word "Biological" on my degree... but will I be a real engineer?

A lot of synthetic biologists come into the field from computer science or electrical or civil engineering, and the whole point of synthetic biology is to turn biology into a Real Engineering Discipline that systematically uses ideas like redundancy in design. If you count up the most prominent synthetic biologists and their origins, it's easy to get the idea that "foreigners" from Real Engineering Disciplines are coming into biology, understanding the science within a week or two, and then dragging it kicking and screaming out of laboratories into design studios and factories. This, too, bothers me.

I do believe that this endeavor will be a truly collaborative one, and will require people with a thorough grounding in biology-the-science as well as people with classical engineering training. I even have a couple of professors' opinions to back this up, although I won't get actual data to confirm or deny this belief until I'm at least in grad school. Certainly biological training helps with designing and conducting experiments; you have to know what a Western blot does in order to know when it's appropriate to do one, and to do it properly. But this makes the biologists sound like the servant underclass in the making of synthetic biology.

Even apart from all this, I worry that synthetic biology, as the newest addition to the family of bioengineering subfields, is not yet ready for its practitioners to be trained in bioengineering instead of in biology or in engineering. I worry that, with an interdisciplinary education, the only thing I'm becoming good at is dabbling, and that I won't develop a true expertise in any particular subfield. -- Then again, isn't undergrad for exploring, and grad school for developing a focus?