• Google
    This Blog Web

October 2011

Sun Mon Tue Wed Thu Fri Sat
            1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

RSS Feed

Bookmark and Share

Email Feed



  • Powered by FeedBlitz

« Fast Takeoff: RepRap Rocks | Main | What I've Been Up To: Risk Analysis »

May 01, 2009

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Hervé Musseau

Was there any doubt that we could do better design than what nature came up with through randomness and natural selection? Of course it took a lot of time, and it may take us some time too to replicate function and improve on performance, but there is no doubt in my mind that (barring some catastrophic end of humanity) our directed designs will out-perform nature's evolved designs.

Erin

One possibility is to bond/add metal atoms and other atoms to the proteins, to stabilize and strengthen them, but, would that interfer with their foldability?

graphene

with reference to your linked article about entropic springs, I have an observation: you say that an entropic spring will push back when compressed while I don't know that's the case. Your example of an elastic band is an excellent analogy; You are correct when you point out that the force with which it pulls back when extendend is entropic in nature, but when you bring the two ends close together it doesn't push back; it sags. I don't think this is any different on the nanometer scale.
Also, concerning possible speed increases due to more rigid springs, the next bond more rigid than those found in proteins would be covalent bonds, and when you start using those as springs, you're just doing IR spectroscopy. I don't think these higher reaction speeds you mention are attainable; I think it would be hard to ship reagent molecules to and product molecules from the reaction site at those rates (fraction of nanosecond per reaction).

Chris Phoenix

Erin, yes, proteins already use metals in some cases to fold with and stabilize. Look up "zinc finger."

Graphene, if you squeezed a rubber band cross-wise (like a gasket) it would certainly push back. Entropic springs like to be in spheres, and whether pushed or pulled, the sphere will be deformed and try to return to a more low-energy shape.

I don't have my notes handy, but I don't think GHz reaction rates have been seriously proposed for any MM design, or are required. As far as shipping product molecules, diffusion is certainly slower than we want, but mechanical transport can be a lot faster.

Chris

graphene

I remain unconvinced on the entropic springs. The onus is still on you to explain why an elastic band doesn't push back when compressed lengthwise. I will try to explain why I think it pushes back when compressed crosswise.
Consider a sphere of elasic band material. When one spot on the surface is fixed and the opposite end is pulled away, it pulls back due to an entropic force; I think we agree on that.
When the opposite end is pushed toward the fixed spot however, the sphere will become bloated around the middle, indicating that material at and around the surface is being stretched. It will pull back on itself, trying to counteract the bloating, and causing the force that seems to push back.
The point is that the entire glob of material is needed to produce a "pushing back" force, whereas only an arbirarily thin line of material is needed to produce a "pulling back" force, in the case of extension.
In the case of an elastic band being compressed crosswise, I think this is the effect you are seeing.
In the case of the two ends of a long, narrow molecule being brought together, there is no surrounding material, so I don't see which force would push them back apart. I hope I have made my point clearly.

I realise this is going a bit off topic so feel free to delete this post from your blog and continue the discussion via email..

Chris Phoenix

Graphene, you're thinking about bulk material. That bulk material has different properties from the individual molecules. A rubber band stays long and thin, in the shape it was molded. A single isolated molecule of rubber will not stay long and thin, but will curl up into a blob.

In a sense, then, a single molecule of rubber (whatever the shape of the rubber object that it's a part of) will behave like the solid sphere of rubber that you imagine. If you pull opposite points apart, the middle becomes thinner. If you push the points together, the middle becomes thicker. Either way, it's out of shape, and wants to return to a sphere.

A way to check your intuition about compressing a rubber band lengthwise is to imagine a series of rubber bands, gradually getting thicker and shorter, until you get a cube. The cube will certainly push back if compressed "lengthwise." Short thick bands will also push back, somewhat. A sufficiently thin band will be very flimsy, and will not push back a lot... but it will push back a little. Wad up a rubber band, throw it in the air, and it'll form a circle before it hits the ground.

BTW, I don't delete blog posts unless they're spam or wildly unconstructive.

Chris

jim moore

Is it correct to think of an entropic spring as being stretched or compressed? Shouldn’t it be thought of as moving from ordered to disordered? I thought that when an entropic spring absorbs energy as it moves to a more ordered state (as opposed to being stretched or compressed) and when it moves back to a more disordered state it returns some (most?) of the energy.

Chris Phoenix

An entropic spring, when in its relaxed (roughly spherical) state, is most disordered. When forced into a different configuration (or, more precisely, a different configuration space in which to wiggle around), it becomes more ordered. This requires energy.

Forcing the spring into a different configuration space can be done by mechanical constraint, such as closing steric barriers around it ("compressing" it) or pulling on the ends of the molecule ("stretching" it).

If mechanical motion requires energy, then a force is apparent. So it's at least approximately correct to think of stretching and compressing entropic springs. As long as you don't move too quickly, and are aware of the relationship between mechanical configuration and spring energy, you can design machines that way.

If the configuration space is changed quickly, energy will be lost. If it is changed slowly, energy will be conserved, analogous to isothermal compression of a gas. This is why efficient protein machines that use entropic springs are speed-limited.

Chris

todd andersen

anyone out there :)

The comments to this entry are closed.