Studying Molecular Biology

An automatic transmission is very complicated. Check out this picture of one component. Or this picture showing the number of components.

Of course, transmissions were designed by engineers, so we know what they are for and how they work. But if a future civilization found a single, broken-up, rusted-out transmission that had been underwater for a couple thousand years, would they be able to figure it out? Probably not.

And if that was the only example of mechanical equipment available to a future civilization trying to reinvent an industrial base, would they conclude that mechanical devices were all mind-numbingly incomprehensible? Perhaps so. And that would be a mistake. Complexity and intricacy are inherent in devices, but comprehensibility is a function of the information available.

Here is a real-world example of such a puzzle: the Antikythera Mechanism which dates from 100-150 BC and was found in a shipwreck. Until they used computerized X-ray scanning to read hidden inscriptions on the gears, no one could figure out exactly what it did, even though they knew it was some kind of astronomical calculator.

For another example, check out these photos of a mechanical analog computer. Check out the pile of parts on the lower-right corner of the page. Now imagine trying to reconstruct the computer... if half the parts were missing.

Now, suppose our future investigators had an unlimited supply of automatic transmissions to study, but the only way to open them up was to drop them off a cliff, and then for some reason they had to examine the resulting fragments by feel, without even being able to look at them. Fortunately, if they managed to damage them just right, they could still turn the shafts and detect more or less what happened with random pieces missing. That is approximately the state of molecular biology today. The good news is that recent tools such as RNA interference let them remove individual "parts" without damaging the rest of the mechanism--woohoo, what a breakthrough!

If you haven't done it yet, go back and click on that first link. It shows a metal plate with dozens of channels in it. If you could only detect those channels by touch, and you had no idea what they were for, you might give them names like L24U19 to describe their twists and turns. After intense study, you'd be able to give some of them more meaningful names and association: you'd probably figure out that L24U19 was connected to the ThroughPlate Channel, which was probably important because that channel goes to the Vanes, which are connected directly to the Main Shaft. The more you learned, the more complex the mechanism would appear. Before long, only specialists would be able even to read the reports of the investigation.

Here's an example of the state of the art in molecular biology. This is a semi-random paragraph taken from the ScienceSampler blog. The purpose of that blog is to report on cell-science papers in more-understandable language: "We read them so you don't have to!!!"

Doa10p is an E3 ligase, that is it is the enzyme that attaches ubiquitin to the protein destined to be degraded. It's also a membrane bound ubiquitin ligase that is involved in the extraction of endoplasmic reticulum (ER) proteins that have a misfolded cytoplasmic domain (i.e. ERAD-C, see this note on ERAD and this note on the 3 types of ERAD). Previously Doa10p had also been implicated in the degradation of mat-alpha2 transcription factor, a nuclear protein involved in turning on genes in response to mating factor. So what's happening?

Eventually, as our tools get better, and as we learn to be more precise and less destructive in opening up the cells and detecting the mechanisms, we will gain a full understanding of how cells work. And we will surely learn techniques that we can use in engineered nanoscale machines. But to date, we don't even know what everything in a cell is for, much less how it works. The known functions of RNA have multiplied in recent years: see this post for a brief overview. And my impression is that the construction and function of carbohydrate molecules have barely been looked at yet; I bet they do at least one important thing that no one has suspected yet.

Eventually, as I say, we will have a picture of a cell that is as detailed and accurate as our pictures of automatic transmissions. A really good mechanical engineer who'd never seen a transmission could probably figure out how it worked by studying its mechanical structure in complete and accurate detail. It wouldn't be easy even then, but it would be possible. And when we know how each part of a cell works, when we can name and classify each part and combine them into systems, we will be able to construct overviews that summarize the functions in far simpler terms.

Meanwhile, studying complicated systems for whatever inspirations we can glean is a very good idea. But it would be a mistake to assume that every useful nanoscale system must be that complicated, or must use the same techniques. While one set of researchers is breaking their brains over whether Doa10p really is implicated in the degradation of mat-alpha2 transcription factor, another set of researchers can be taking simple concepts--the "gears" and "clutches" and "pumps" of the cellular world--and making useful things with them. Even simpler concepts--"levers" and "springs", by analogy--can be used to make useful machines including calculating devices.

So I'll close with some questions. At what point will we know enough to start conceptualizing useful devices based on simple aspects of the nanoscale--devices we might even want to try to build? Is that point likely to be reached before, or after, cell researchers have managed to figure out how cells work? Could that point have been reached already?

Chris Phoenix

Tags: nanotechnology nanotech nano science technology weblog blog

Comments

[Phillip Huggan]

I'll dryly add that in the above thought experiment there is no such thing as a good Mechanical Engineer that has never seen a transmission before.

"At what point will we know enough to start conceptualizing useful devices based on simple aspects of the nanoscale--devices we might even want to try to build?"

I'll say various diamond Density Function Theory sims (both inside and outside of MNT circles) can allow us to posit some simple parts. More sims mean a greater library of potential mechanosynthetic operations. Ultimately someone will have to draft an atomically precise nanofac blueprint if a surface sciences chemist is to built it. Drexler's electrostatic motor or some other scaleable actuator will be needed for this whole exercise to work.

[Chris Phoenix, CRN]

Yes, two significant roadblocks at the moment are lack of fast powerful addressable nanoscale actuators, and lack of sims for mechanosynthesis.

Lack of sims will be addressed by funding--and the funding required will fall as fast as computers improve--maybe with an extra decrease now that there are compute clouds for sale.

Actuators... Here's a crazy idea. Push on pads with a scanning probe microscope! You can even get force feedback from an AFM. That only lets you actuate one thing at a time, so you'd need some kind of position-holding mechanism... which is a pretty complex mechanical device... so this probably isn't the best way to do it. Another possibility may be to use long buckytubes to link to micro-scale actuators. Ultimately, of course, you want electrostatic actuators (or possibly electron-shift chemical actuators) with wires to the outside.

Chris