Sunday, February 7, 2016

February 07, 2016 at 05:06AM

Today I Learned: 1) ...this quick algorithm for cooking dumplings: bring water to a boil; add dumplings; for x = 1:3 let water come to a boil; add a cup of cold water to cool; end for let water come to a boil; remove from water and serve; 2) When making spring rolls, don't despair if they don't get wrapped well, or if they crack a bit, or if they come loose. Fry them carefully and they'll be fine. Maybe. 3) ...how to make a binary counter out of genetic repressors! According to the standard simplified repressor model, at least.

Saturday, February 6, 2016

February 06, 2016 at 04:12AM

Today I Learned: 1) ...how to use a Buck converter! That's a circuit component designed to take an input voltage and convert it down to a smaller voltage. Very handy for, oh, say, finding the voltage potential required to evolve oxygen off a catalyst.... This was particularly difficult because I and the high school students involved with this project didn't have any experience with or instructions for a Buck converter, so we had to pretty much guess our way through it. 2) When E. coli get stressed, particularly from starvation, they sometimes transition to an absolutely ridiculous morphology where they keep growing but more or less stop dividing, so they get absurdly long. If you're not familiar with what E. coli normally look like, go do a quick Google image search. Then check out Figure 1 from this paper, in which the authors induced filamentous growth with platinum salts: http://ift.tt/1PqC0xe. Also check out figure 3, where they show what hapens to the filaments when they're moved to platinumless media -- the whole filament shards itself as it apparently undergoes all of its divisions at once. 3) Jonathan Blow, the indie game designer responsible for Braid and, more recently, The Witness, has a fantastic website. It is ancient and ugly, but it has some of the highest density of interesting content I've ever seen on the web. Some of it I've seen before ("The Unreasonable Effectiveness of Mathematics in the Natural Sciences"), most of it I haven't ("The Unreasonable Effectiveness of Mathematics"*). I have not agreed 100% with anything on the site yet, but just about everything has given me a new way of looking at something I'd thought a lot about before. Speaking of which, it's time for me to go into unpaid advertisement mode -- The Witness is really good. Really, really good. I'm not done yet, but if it's going where I think it's going, it's one of the best-executed pieces of psychological manipulation I've seen in the medium. *Not a typo.

Thursday, February 4, 2016

February 05, 2016 at 02:26AM

Today I Learned: 1) Anders Knight is on a roll recently -- tonight, he is responsible for information on the Wendelstein (pronounced "vendelshtein"), the Max Plank Institute for Plasma Physics' brand new fusion power research facility. You can get a lot of information here (http://ift.tt/1VTi8GW), but I'll relay some of it anyway. The Wendelstein is a fusion reactor of the stellarator type (for a comparison of stellarators to the more common tokamak designs, see http://ift.tt/1SwHUD7. This is also a really good fast introduction to how a fusion reactor works, in its broadest strokes). It will not produce power. It is instead designed as a research facility to test different configurations of magnetic containment* so we can figure out how to optimally build fusion reactors that might, in a few decades, produce power. One of the cool things about the Wendelstein is how long it can operate -- the facility has already performed runs as long as a quarter second, and plan to ramp up to thirty seconds. For a fusion reactor, this is an eternity. Another cool thing about the Wendelstein is the insane exactness required to build it. For example, each of the 50 non-planar (I'm not sure what that means) superconducting coils of the Wendelstein's magnetic containment system measures roughly 3.5m x 2.5m x 1.5m and is machined to a precision of *three millimeters*. Speaking of the magnetic containment system, another cool thing about the Wendelstein is how *awesome* it looks. I mean, check this out: http://ift.tt/1VTi8GY. That's not science fiction, that's engineering that's being done *right now*. * Fusion reactions only happen at super-high temperatures, temperatures which are hot enough to destroy literally any known physical material you put them in. Fusion reactors contain their plasma with complex magnetic traps. You know in Star Trek when ships lose magnetic containment and explode? That's because the plasma in their fusion reactors escapes the reactor and briefly exposes the rest of the ship to temperatures not dissimilar to the interior of the sun. 2) When E. coli cultures are infected with bacteriophages (viruses that attack bacteria), it apparently often causes the culture to form visible clumps or filaments. There's also some rumor that phage-infected cultures tend to get really foamy, from all the protein released by dying cells. 3) In a drug (or medical device) trial, each subject costs, on average, about $10,000. The total cost of passing regulation for a medical device averages around $150 million. Source: a lecture by a senior guy from Medtronic.

Wednesday, February 3, 2016

February 04, 2016 at 02:31AM

Today I Learned: 1) Big news! http://ift.tt/1VOYCLF 2) Something to speed up PCR linearizations, or really any amplification you have to check on a gel -- start DNA purification while the gel runs, before you know what needs to be purified and what doesn't. It'll cost more, because you end up prepping some samples that didn't turn out anyway, but it saves some time. I'm pretty sure I knew this at one point (iGEM, can you confirm?), but I learned it anew today. 3) Hexane is a surprisingly cheap solvent. It also comes in a LOT of purities: http://ift.tt/1JZCX2F

Tuesday, February 2, 2016

February 03, 2016 at 02:12AM

Today I Learned: 1) Let me tell you the story of FDC Willard, one of the most famous published authors in low-temperature physics. The story of FDC Willard is inextricably bound to the story of Jack Hetherington, a low-temperature physicsist. In 1975, he submitted a paper of his for review, which happened to be a single-author paper. He learned from a friend of his during reviews that his paper would be rejected on editorial grounds because he had used the standard "we" throughout the paper, which was in this case improper as there was only one author. Jack didn't want to completely re-work the paper, so he decided to add a second author instead. Apparently having no colleagues in low-temperature physics who he wanted to add to his paper, he decided instead to add his cat, Felix domesticus cattus Willard, i.e. FDC Willard. The paper was accepted. Willard would later go on to first-author a paper on... well, frankly I couldn't decipher what it was about, but it was another low-temperature physics paper having something to do with helium 3. Thanks to Anders Knight on this one! 2) Enzymes are really useful for a lot of industrial chemistry. It's relatively common to mix up a big reaction vessel with some substrate, pour in a bunch of enzyme, and stir it around or heat it to produce a reaction, and the enzyme greatly reduces the energy required to do so. The product can be extracted, sold, and thus money is made. However, enzymes are expensive to produce. You really don't want to make a giant reaction vessel's worth of enzyme and then throw it away. You can filter out enzymes with some sort of antibodies or similar trick, but that tends to be expensive and inefficient. A generally better way to re-use enzymes (as I learned today) is to not just dump the enzymes in -- instead, covalently bond them to some solid thing, like a rod or paddle or mixing aparatus, which you can stick in the reaction vessel... and just pull out when you're done, ready to be re-used. Enzymes used this way are referred to as immobilized biocatalists. The tricky part is figuring how to attach the enzyme without destroying its function. Rumor has it that one surprisingly effective way to attach enzymes to a surface is... to bake them on! That's it. Just heat them a bunch. Apparently works better than it should, although I don't have a reference for that. Credit also goes to Anders Knights on this one. 3) Otto E. Rössler, inventor of the Rössler atractor (a rather famous system of simple differential equations with rather complex properties that's studied as a cannonical example of chaotic dynamics -- check out the wiki page for some cool snapshots of the system) is one of the most wide-ranging scientists I've ever encountered, and possibly one of the craziest. For example, he is one of the scientists most against the experiments in the Large Hadron Collider on the grounds that it could produce Earth-destroying black holes. Check out his blog, linked below. Among other things, it covers: cryodynamics (a "sister theory to thermodynamics"); ebola; the LHC; Interstellar; the ethics of curing elephant autism. http://ift.tt/1mc997p

February 02, 2016 at 03:53AM

Today I Learned: 1) Our best images of single stars (other than our own) come from IR interferometer telescopes. IR means it collects infra-red light, presumably because it... actually, I don't know why IR would be better than visible light. In any case, an interferometer is a telescope array that collects the same image and compares them from slightly different angles*. Anyway, one of the IR interferometers has captured some pictures of other stars that are, like, more than one pixel! We're talking about images on the order of a couple dozen pixels square. It turns out that a lot of the big ones (which are easiest to image) are really squashed-looking -- they rotate so fast that they bulge at the center. This effect happens in all spinning bodies, but on the ones in our solar system the effect is small. *Apparently the total resolution you get is the sum of the resolutions of the individual telescopes, and it's easier to build a bunch of small telescopes than it is to build one big one. You can probably also split up an array to do a bunch of low-resolution tasks at once, though this is just speculation on my part. 2) Glassblowing, as a field, has a lot of really archaic verbage. Here are a couple of terms I learned today (I learned them verbally so the spelling is virtually guaranteed to be wrong). The puntee is the rod on which you gather glass. Puntees can be solid, for making solid glass pieces, or hollow, so you can blow into them to create things with holes or big rounded cavities. Gathering glass, by the way, is what you do when you put a blob of glass on the end of a puntee. The blob of glass on the end of the puntee is the moil. One of the common things one does with a moil is to roll it on a mauvre, which is a plate traditionally made of marble (from which the name comes), but now is more likely... steel? I'm fuzzy on this. In any case, you roll a moil around on the mauvre to shape it. Thanks to Tasha Johnson for teaching me these! 3) ...what a Luer lock is. Basically, it's a standard of screw fitting for plastic parts like syringe tips and tubing. I've been using them for a while now but I just now learned what they're called.

Monday, February 1, 2016

February 01, 2016 at 03:01AM

Today I Learned: 1) So thalidamide is this decently effective drug whose enantiomer is highly teratogenic, which led to tens of thousands of severe birth defects when the drug was first administered as a painkiller for pregnant women(!). It's always struck me that this was an awfully sloppy mistake to make. After all, chirality is something chemists know about, and presumably it would have come up during the synthesis of thalidamide that there was an enantiomer they should really deal with. Wy did anyone think it was a good idea to administer both isomers of a drug? Well, it turns out that it wasn't just sloppy chemistry -- thalidamide can *switch its own racemization*. If you make a pure mix of 100% of either thalidamide enantiomer, it will decay back to a 50/50 over time, *especially* in biological conditions. I had no idea there were compounds that could do that. (I'm actually not sure whether the original thalidamide problem was because of this, or because the chemists that made it were sloppy. This bears more research) 2) Here's a bit of a thought experiment I've been thinking about for a while. Take a box full of salt water (say, sodium chloride). Apply an electric field across the water by putting a positively-charged rod at one end and a negatively charged rod at the other. The sodium ions should be attracted to the negative pole, and the chloride ions should be attracted to a negative pole, so this will create a gradient of charged ions across the box. Now spoon out a bunch of the water at one end, near the pole, and seal it away in a jar. Do you now have a bunch of charged water? It seems like you really, really ought to, but the idea of water with a net charge seems really crazy to me... ...well, today I was reminded that we make charged water for a bunch of reasons using a technique called electrospray. There, though, you're putting charge on a bunch of water droplets and spitting them out using an electromagnet. It turns out that if you put TOO much charge on a droplet, the repulsive forces of the charge overcome the cohesive forces of the water and it explodes. There's a special name for this explosion -- a Coulomb explosion. 3) ...not to let the lab turbidostat sit on its own for more than a day, or it starts to spew slightly contaminated media all over the bench. My secondary containment system (a box) worked decently well at limiting the damage, at least....