July 31, 2016 at 11:33PM

Today I Learned: 1) Did you know you have a blind spot in each eye? Did you know it isn't in the center of your eye? I didn't. Here's how you can find our own blind spot. You can do it right now. Get a piece of paper and a marker. Draw two dots, about a half inch wide each, about a hands-width apart. Hold the paper in front of you at a short arm's width, with the dots forming a horizontal line. Close one eye. Focus on the dot on the same side of your face as the closed eye (if you close your left eye, focus on the left dot, and vice versa). Now slowly move the paper toward yourself, keeping your gaze focused on the dot. At some point, the other dot will disappear from sight, then reappear. It's a pretty small spot, so you'll need to move slowly. 2) Acid rain has apparently gotten rather better since the 80s and 90s, when it was really big in the public eye. At least, in America it has. In 1992, average rain pH on the east coast of the US was between 4 and 5 -- around the acidity of apple cider vinegar. Moreover, you can tell whether a rain storm on the east coast came from the Atlantic or from inland by the pH -- Atlantic storms have a pH betwen 5 and 6, while inland acid rain is closer to 4. Since then, it looks like environmental regulations and cap-and-trade limits on sulfide and nitrate emissions have brought acid rain levels down considerably. SO2 emissions have been successfully reduced to levels mandated by late 80s legislation *ahead of schedule*. This, along with the fluorochlorocarbon ban in 1987, serves as a reminder that government intervention *can* mitigate or even reverse severe environmental damage caused by gas emissions, as long as there's the political capital to put necessary measures into place. (Now, acid rain in *China* is another matter: http://ift.tt/2aHOvvB) 3) Ever wonder why we have mines? To have a mine, you have to have a big old pocket of some valuable metal, in one spot, easily accessible (relatively speaking). Why are there deposits of metal? Why isn't all of the rock on Earth a well-mixed, homogeneous mass? Well, for the most part, it is -- lots of silica and iron and magnesium and some other metals, mixed up and spread around. Metal deposits typically happen when 1) some event brings up metals from the mantle and 2) some kind of formation exists to transport those metals to 3) a formation that can trap or accumulate those metals. Take copper ore, for example. Copper ores are usually formed from volcanic activity or some other geologic activity that brings a lot of molten magma up from the core. Often the magma ends up running through underground tunnels, where it starts to cool. The magma doesn't actually have much copper in it -- it's mostly silica and iron, like the rest of the crust. As the outside of the magma flow cools, it crystallizes into iron-silica rock crystals. Copper ions don't fit well into iron-silica rock crystals, so they're excluded from that crystallization process. As the outside of the magma tube cools and crystallizes, it concentrates copper near the center of the tube, like salt concentrating in an evaporating tidal pool. Eventually the whole thing freezes, leaving a rich vein of copper metal (or,at the very least, copper-rich rock) at the center. Bonus rock fact: One of the unusual housing hazards is radon buildup. Radon is a radioactive gas with a half-life of about four days. It's an alpha emitter, which means it spits out high-energy protons when it decays, which are very damaging to biological tissue if it gets under the skin. It also tends to leach out of bedrock, and can accumulate in houses built on top of radon-rich ground. Wait a second, if radon has a half-life of four days, how is there any left in the ground? Turns out radon is a decay product of *uranium*, which is common enough in some soils to produce measurable amounts of radium, albeit very slowly. It's not a problem normally, but if a house is built in a way that acts as a radon sink... that's a problem.

July 30, 2016 at 11:53PM

Today I Learned: 1) ...what tonic water tastes like. Firstly, I should note that American commercial tonic water has more high fructose corn syrup and citric acid than quinine, so tonic water has a soda-like, fruity-sweet taste. Underneath that, though, is a bitterness that adds a surprising amount of depth. Overall, I was pleasantly surprised by the taste of tonic water. 2) The human body is not quite symmetric -- humans have their hearts on the left side of the chest, and their liver on the right side of the abdomen, among other things. About 0.01% of the population has their organ asymmetry reversed, with the heart on the right, liver on the left, etc. Today I learned that this condition is called situs inversus, and is mostly medically irrelevant unless a doctor has to operate on you, in which case they really need to know what's up before they cut. I also learned that some, but *not* all(!) of the asymmetries in the *brain* are also reversed in situs inversus. This, I find truly puzzling. 3) There is a (prototype) wearable unicorn horn that takes videos of whatever the wearer is interested in. The horn has an EEG sensor that monitors the user's attentiveness. It detects when the wearer is focused on something, and when that happens, it starts taking video. The main goal is to use this on kids with ADHD and other attention-related difficulties, allowing researchers or teachers or parents to see what it is that gets the kids' attentions. It could also be used as a sort of passive, selective memory-recording device, like a 24/7 recording cam but selective only for the things you actually pay attention to. The horn's artistic designer, Anouk Wipprecht, has made some other amazing prototype wearables. She designed a drink-pouring dress which will dispense drinks only to people who retain a respectful distance (if you play a phone-based Truth or Dare game to the host's satisfaction, it can then add alcohol). She also designed a gorgeous dress/exoskeleton with spider-like arms on the shoulders, which sense approaching persons and will either extend aggressively or beckon welcomingly depending on the wearer's detected mood. She *also* created a somewhat-less-stylish wearable full-body Faraday cage. This is some seriously cool tech -- here's hoping that kind of wearable technology becomes commercially available.

July 30, 2016 at 12:05PM

Today I Learned: 1) Tonic water is water with quinine. Originally, tonic water was consumed for its anti-malarial benefits, but it tasted so bad that Englishmen eventually started adding sugar and carbonation for taste, which eventually led to gin and tonic. Modern tonic water has considerably less quinine (about 300 µM) than original formulations (not sure how much it used to be). 2) Quinine in water (so, tonic water) glows bright blue under UV. Someday, I should get some tonic water for testing "UV flashlights" and blacklights (eggshells also work, but as a vegan, I'm less inclined to get those...). 3) There are sunglasses for dogs. They are called "doggles".

July 29, 2016 at 01:48AM

Today I Learned: 1) Spider silk products might just happen after all. For those not in the know, spider silk is a bit of a nanotechnologist's dream. It's smooth and silky like, well, like silk, but it's also super elastic and has a ridiculously high tensile strength, much higher than steel and appreciatly higher than kevlar. Moreover, different kinds of spider silk, made by mixing different protein subunits in different ratios and with different spinning techinques, can have different properties. Personally, I count spider silk as my version of flying cars -- when we have commercially-available spider silk, we will officially be living in the future. Again. Why don't we just farm spiders? Well, the trouble is that spiders are really hard to cultivte. Notoriously hard. The main problem is that just about all spider species are cannibalistic, so they can't be grown in very high density. That hasn't stopped some people from harvesting spider silk for select applications, but it's damned expensive (~$100,000/kg, according to Kraig Biocraft Laboratories, which I'll get to in a second). Scientists have been trying to reproduce spider silk for a long time, and more recently, they've been trying to produce it sans spider. Enter Kraig Biocraft Laboratories and Bolt Threads, two relatively new companies looking to put spider silk on the open market. I'll start with Kraig Biocraft Labs. Their approach is straight-up biotech -- they're experts with silk protein genes, and their goal is to bioengineer silkworms to produce spider silk instead of silkworm silk. If that works, then KBL can just slot the spidersilkworms into a standard silk-harvesting operation and they'll have bolts and bolts of spider silk. While that's really cool, the one *I'm* actually rooting for is Bolt Threads, a *very* recent startup that's taking a more mixed approach. A lot of their techniques are hush-hush right now, but what from what I gather, they're producing silk proteins en-masse in bioengineered yeast, extracting the individual proteins (not easy! Spider silk likes to crash out of solution and form thread, so storing it in liquid form isn't trivial), and physically weaving them into thread using robots (microfluidics?) inspired by spider spinnerettes. This requires a *lot* more new technology than KBL's plan, but has the advantage that Bolt Threads can quickly pivot into different silk versions, with extreme flexibility in the properties of the silk they produce. At this stage, Bolt Threads is also marketing very heavily on the idea that their thread is much cleaner to produce -- since it's produced straight from purified protein to thread, it shouldn't require the harsh, environmentally damaging chemical treatments that traditional textile industries use to get the kind of quality we expect from textiles these days. 2) It is possible to sing underwater. There's a group called AquaSonic that has perfected the art of underwater music, including underwater singing. Their music includes, among other things, vocals, gongs, hurdy-gurdies, violins, chimes, and aquarium walls. Check them out: http://ift.tt/2a8MvHN Thanks to Sarah Seid for bringing this to my attention! 3) There is at least one species of orchid whose roots smell truly vile. At first I thought it must be some sort of means of attracting pollinators, but if that's the case... why the roots?

July 28, 2016 at 04:23AM

Today I Learned: 1) ...how to use the software for the Victor plate reader. It's an older model, but it works. 2) When 3D-printing, it makes sense to have the printer head move slowly along the outside edge, and much faster on the infill*. The more slowly the head moves, the smoother the surface it will make, but it would take a lot longer to run everything at lowest speed, and the smoothness of the interior of a structure doesn't matter. Also, typcial infill for 3D-printed structures that don't have to be super tough can be quite low, like 10% low. * Infill is the material printed between outer walls. You *could* print objects to be totally solid, but that would waste a lot of material; similarly, you *could* print objects using only their outlines, but that would be quite flimsy. Infill is the stuff you use to fill between the outer lines, which is usually some sort of honeycombing pattern. 3) You can switch Cas9 from being a cutting protein to a DNA-binding-only protein (i.e., an activator or repressor) by switching its guide from a 20 bp RNA to a 14 bp RNA. I'm not sure how much I would trust this technique... but it worked for Kiani lab at Arizona State University, according to this abstract: http://ift.tt/2ac9MJP.

July 27, 2016 at 03:18AM

Today I Learned: 1) Dogs are much more prone to giving up on problems and asking for help than wolves are. If you give a wolf a difficult problem, like getting into a cage to get at food, it will go at it for quite a while, trying different avenues of attack, generally not giving up. Most dog breeds will try it a few times, then give up and effectively ask their owners for help. 2) ...how to laser cut. Not really one big thing to say here, just a lot of little things, like how to make a line actually red in CorelDraw software, and the difference between material-based settings and custom settings, and how to make sure a cut will stay on your material. Also, how to cut acrylic with a hack saw (it's a lot like cutting anything else with a hack saw -- in particular, it's helpful to clamp down as much of the material as possible and to hold the other end steady with your free hand). ...also, if you walk into a lab confidently enough and ask for help using the laser cutter, you will at least sometimes get exactly the help you need. =) 3) Topical treatments for toenail fungus are usually paired with chemical debridement, which is fancy doctor-speak for dissolving away unwanted toenail with enzymes. Debridement helps the topical treatment get to the fungus, which is the primary problem with using topical treatments on toenail funguses.

July 26, 2016 at 03:34AM

Today I Learned: 1) ...the difference between TCP and UDP. Both are protocols for transmitting data over the internet. UDP (User Datagram Protocol) is a lightweight, relatively simple protocol with a simple header and not a lot else. It's cheap to set up, super fast, and totally un-guaranteed to work (it's a "best effort" protocol). TCP (Transmission Control Protocol) is a much more heavyweight protocol that emphasizes security and transmission reliability over raw speed. In TCP, two communicating entities set up a channel with a sender on one end and a receiver on the other. The sender sends out packets over this channel, and the receiver transmits back a receipt of each packet to let the sender know that the packet made it through. The sender will wait until it gets the receipt, and if it doesn't get such a receipt (in the event of packet loss of either original packet or the receipt packet) it will re-send the packet until it does work. This type of communication makes TCP super slow relative to UDP. For many applications (i.e., email), that's fine -- data accuracy is much more important than data speed, in those applications. For data-heavy or time-sensitive communication, as in voice chat, UDP is better because it is less prone to terrible slowdowns. 2) The scientific name of the llama is "Lama glama". I can't get over this fact. I just can't. 3) England has a power station, the Dinorwig power station, that effectively exists entirely to run the nation's tea kettles. See here: http://ift.tt/2amJRkq. The Dinorwig power station is actually a giant hydroelectric battery. When lots of power is available on the grid, Dinorwig pumps water into an elevated reservoire. When additional power is required, Dinorwig opens up sluices on the water reservoires and uses them to spin its turbines. Why? Dinorwig was actually built to solve an issue with nuclear plants, back when the UK was poised to build a lot of nuclear power plants. Nuclear plants apparently can't really run at much less than peak output, which means that a lot of the time they'll be wasting lots of power. Stations like Dinorwig can store this power for use during spikes, or can let nuclear plants shut down for a while to increase efficiency. Much of the UK's nuclear plans didn't hatch (although something like a sixth of England is powered by nuclear these days), so Dinorwig doesn't really serve its original purpose. It's still good for handling power usage spikes, though. The most common cause of such spikes is when a major network program goes to advertising, which can prompt a significant fraction of England's population to turn on their water kettles at once. Thus, one of the most important factors in Dinorwig's operation is the nation's network broadcasting schedule. Thanks to Michael LeFew for pointing me to this.

July 25, 2016 at 02:52AM

Today I Learned: 1) Resistor values are sometimes written in the rather odd notation aXb, which means "a.b, multiplied by some power of 10 designated by X". It's clearer with an exmple -- 1k5, for instance, means "1.5 thousand" and 4M2 means "4.2 million". Why would resistor values be written that way? First of all, resistor values can span lots of orders of magnitude, which makes scientific notation a natural format for writing resistor values, i.e 1.5 x 10^3. However, in small fonts (and especilly, I suspect, early printed fonts) dots can be difficult to see, so it might be hard to read whether a resistor is 1.5 kilo-ohms or 15 kilo-ohms. The notation "1k5" makes it quite explicit where the decimal goes. Nobody's going to mistake 1K5 for 15K. Unless you're dyslexic. Then all bets are off. 2) MRSA is not, as I previously thought, a *multiple* resistant bacteria -- it's a *methicillin*-resistant bacteria (methicillin-resistant Staphylococcus aureus, to be precise). Methicillin is an antibiotic related penicillin and ampicillin (a β-lactam antibiotic) that interferes with cell wall synthesis. MRSA is any strain of S. aureus with resistance to methicillin (and other β-lactams). Resistance is granted by a plasmid carrying, among other things, the mecA gene, which encodes a protein that binds to β-lactams and competitively keeps them from messing up cell wall synthesis. Interestingly, the name MRSA is a bit anachronistic, as methicillin is apparently almost never actually used anymore, even in lab testing. Other β-lactams have fewer side effects, better potency, or better shelf life. 3) I was poking around the innards of an xbox one controller today, and I was wondering why they use the annoying little security screws -- they use a hex head with a little nub in the middle that blocks any screwdriver without a hole in the middle, so you have to get a special screwdriver to open them. Sure, they might be a little more "secure" against tampering, but let's be honest -- if anyone really wants to get inside an xbox controller, they can just order the right screwdriver on Amazon and they'll have it right open in a couple of days. It certainly doesn't secure the controller against their competitors. So, why do they use it? I decided to go look it up, and while I didn't find out exactly why Microsoft uses those screws on their controllers, but I did learn a bit about the screws and why they're used in general. That style of screw is called a torx screw, and it's primarily designed to not strip when overtightened. To my amazement, it turns out that philips-head and flat-head screws are actually designed *to* strip when over-tightened. I guess the idea is to protect whatever you're screwing into from overtightening damage, but it hardly seems worth making the screw effectively no longer a screw. Anyway, torx screws are designed to not strip no matter what, and so rely on the user to not over-tighten. Maybe they're easier to screw on by robots?

July 24, 2016 at 06:20AM

Today I Learned: 1) ...how to mount a router to a wall, and how to staple cables around the edge of a room. 2) Elderberry juice is really tasty -- it's kind of like a slightly sour white grape juice, with similar rich, deep hints of grapiness, for lack of a better word. 3) A nuclear thermal rocket is a rocket design powered by gas (usually hydrogen) heated by a nuclear reactor. The rapid expansion of the gas fuel as it is heated causes it to shoot out the back, providing thrust. Nuclear thermal rockets haven't really been used in spacecraft, but they are predicted to be pretty expensive and generate less thrust than traditional rockets, but should also be lighter and much more fuel-efficient. Nuclear thermal rockets are not to be confused with nuclear pulse propulsion, which propels a rocket by exploding small nuclear warheads behind a blast-resistant shield.

July 23, 2016 at 05:35AM

Today I Learned: 1) ...what a persian cucumber tastes like. It's a lot like any other cucumber, except with more of that slightly bitter, refreshing cucumber taste. As a bonus, I learned that a bit of lemon juice, salt, and pepper do wonders for a cucumber salad. 2) One of the major classes of antibiotics is the aminoglycosides, which includes kanamycin, neomycin, and streptamycin, among many others (incidentally, antibiotics with "mycin" in their name are derived from Streptomyces bacteria). Aminoglycosides work by binding in the bacterial ribosome, which messes up the ribosome and stops it from making functional proteins. I always assumed that aminoglycosides just stopped protein production, but today I learned that aminoglycosided ribosomes actually continue to make protein... but they don't incorporate amino acids accurately, so the proteins they make usually have the wrong sequence and often are prematurely terminated. As a bonus, today I also learned that aminoglycosides are something of a last-resort antibiotic, or a general-purpose one if the infection can't be diagnosed. They're specific to bacterial ribosomes... but not *completely* specific to bacterial ribosomes, so they can cause some nasty side effects in humans. 3) Transposase libraries are usually made with viral transposons. I'm not sure why viral transposases... but then, I'm not intimately familiar with the protocol for building a transposase library. The basic idea of a transposase library, by the way, is to generate tons and tons of variants of (usually) a bacteria with genes randomly knocked out. Transposases are small genetic elements that can randomly insert themselves into the genome of a host, and can be found in pretty much all clades (though they're much more common in Eukaryotes, especially Eukaryotes with big genomes). To make a transposase library, you introduce a transposase to a population of cells and induce the transposases to insert for a short time. Each cell will get a transposase in a different location. If the transposase inserts inside a gene, it will usually shut down the function of that gene, so the library of transposase-infected cells serves as a knockout library that you can screen in the usual ways.

July 22, 2016 at 02:44AM

Today I Learned: 1) There is a bird that has evolved symbiosis with humans -- the greater honeyguide. Honeyguides will lead humans to bee hives, using and responding to special, locally-varying calls. The human hunters harvest the bee hives, and the honeyguides feast on wax and grubs left over once the people leave. Credit to Mengsha for finding this -- more information behind this paywall: http://ift.tt/29Zmni6 2) You can synchronize lightning bug flashes using LED lights: http://ift.tt/2a4U5GI 3) BsaI restriction enzyme, which I work with frequently, doesn't cut methylated DNA very well, and the hi-fidelity version (BsaI-HF) basically doesn't cut methylated DNA at all! That could explain a fair number of failures of golden gate assembly in our lab over the last couple of weeks....

July 21, 2016 at 02:17AM

Today I Learned: 1) ...how to load (and use) an electric drill. 2) ...of an interesting technique for metabolic engineering. I actually forget the name, but the key is to link the output of a metabolic pathway you want to build to some key growth factor or antitoxin or something else that is good for the host chassis, whether it be bacteria or yeast or something more exotic. Then you basically just grow the stuff. Since the output is good for cell growth, the host will tend to evolve more efficient variations of the metabolic pathway. There are some problems with this technique. The output has to be something either beneficial to the cell or something that can be tied directly to something beneficial to the cell. The more indirect that link, the more likely it is the population will just evolve to break it and just produce whatever it is that is good for growth, dropping the pathway you actually want completely. Still, in the narrow cases where it works at all, this seems like a pretty nifty technique. 3) Saw Hollywood today for the first time. I'm not particularly impressed. It strikes me as similar to the rest of LA, but busier, more densely packed with rich people, and with worse congestion and roadways. There's probably some interesting history to be found actually *inside the buildings* there, but I'm not sure if it's history I care about enough to go back to check out.

July 20, 2016 at 02:17AM

Today I Learned: 1) Today I got to be a cyborg! Specifically, I wore a headset mic with a hip-speaker for a good chunk of the day, for the benefit of those around me whose hearing isn't up to picking up my feeble voice (which is a lot of people). It was a surprisingly fluid experience in terms of audio experience. As I suspect is the case with most cyborg enhancements, the biggest problem was physical discomfort -- I found wearing a mic on my head got annoying surprisngly quickly. It was much more comfortable wrapped around the back of my neck, with the mic pointed up at my face. 2) You can get your genome sequenced for about $1000. Veritas Genetics offers a $1,000 full-genome sequencing service (averaging 30x, according to the company blog). I'm unsure of whether you actually GET YOUR SEQUENCE, though, or if they just give you a bunch of interpretation like 23andme (which runs about $200, and just gives you a bunch of variant-calling, which is cool and much cheaper, but you can't analyze it yourself). FullGenomes offers a guaranteed 30x full-genome service for $1500, and they promise a BAM file, which if you know how to interpret it is all the information you could possibly want. Data analysis services cost more. Then there's µBiome, which will give you a gut microbiota sequence, with analysis, for $89. From what I'm reading in reviews, µBiome doesn't really give you that much useful information, largely because nobody really has any solid idea of what effects different gut bacteria have on the human body. 3) Bio lab fact: There are two definitions of concentration of T4 DNA ligase, cohesive end units and Weiss units, both abbreviated as U (or, more usually, U/mL or U/µL). Cohesive end units are defined as the amount of enzyme required to ligate together 50% of a the HindIII fragments from lambda DNA at a concentration of 300 µg/mL, in a certain buffer, for 30 minutes, at 16°C. Weiss units are defined as "the amount of enzyme required to convert 1 nanomole of [heavy phosphorous]-labeled inorganic pyrophosphate into Norit adsorbable material in 20 minutes at 37°C, using specified reaction conditions", which is about 200 times as much as a cohesive end unit.

July 19, 2016 at 01:04AM

Today I Learned: 1) Today I learned how stomachs get their HCl and enzymes! Yesterday I did not know, but I put the question to the AskScience subreddit, and they delivered! The lining of the stomach is covered in little pores called gastric pits, which are basically infolded pockets of the innermost cell layer (there are some nice basic diagrams here, along with tons of terminology and other information if you're interested: http://ift.tt/2a54TUq). The cells in these pockets secrete a mix of HCl and an inactive protease* called pepsinogen. Under highly acidic conditions, pepsinogen breaks down into pepsin, which is an *active* protease. This way, the cells of the stomach can produce protease safely, without it turning on and shredding the cell from the inside out, and the stomach will naturally convert that into an active protease. The tricky bit is that the lining of the stomach, including the pores, is covered by a layer of mucus that keeps the stomach from digesting itself. How does the acid/protease mix get through the mucus layer, if the whole point of the mucus layer is to stop acid and enzymes from getting past it? The answer is that the gastric pores occasionally contract strongly, pushing the aqueous acid/protease mix into the mucus. It cuts a channel through the mucus and blasts through to the other side. The hole in the mucus heals pretty quickly, and then there's nothing to force the juice back through. The principle behind this was apparently only discovered in the early 90s, as evidenced by this paper (paywall warning): http://ift.tt/2a6ExnI. *protease == a protein that breaks down proteins. 2) I've noticed that my mantis will sometimes bob back and forth when it spots me. I figured it was gaguing its distance to me so it could figure out whether I was predator or prey. Today I learned that the purpose of a mantis's rangefinding isn't to figure out whether you are dangerous or not -- it's to figure out whether it can jump onto you or not. 3) Ikea makes (or possibly made) a rather ingeneious new version of a screw, called a wheel lock: http://ift.tt/2a54HnW. The wheel lock goes into a cylindrical inset in a flat piece of material, with the flat face flush with the face of whatever it's embedded in. A hole in the side of the embedding material lets a pretty standard screw through, which makes contact with the rounded side of the wheel lock. By screwing the wheel lock, you screw the screw into the *side* of the wheel lock, drawing it into place! I didn't know there were novel forms of screw to discover out there -- thanks for making my day, IKEA!

July 18, 2016 at 03:50AM

Today I Learned: 1) Ever wonder how stomach cells don't just get digested like everything else in the stomach? Part of the answer is a mucus membrane, but I think there must be more at play, since the stomach lining can somehow secrete enzymes and acid *into* the stomach, so those must also be able to get *out* to some degree. Today I learned that part of how the stomach keeps its cells alive is by having the stomach lining cell reproduce really quickly. 2) There are soups (including some gazpacho recipes) that include blended bread, digested with a bit of vinegar. Mengsha Gong seemed nonplussed by this fact, so I'm guessing it's well-known among people who *really* cook. 3) ...how to play shogi, the Japanese version of chess. The major differences between chess and shogi: you can place captured pieces (almost) anywhere onto the board as a move; almost all pieces can be upgraded if they reach the other side of the board; the game is played on a 9x9 board instead of an 8x8 board, with the king smack dab in the middle of the first rank. Shogi pieces are surprisingly similar to the pieces of western chess. The king is basically identical in the two games. There are pawns, which can only move forward by one (and capture the same way). There are rooks and bishops, one per side, which move just like their western counterparts. There are knights, which move two spaces forward and one to the left or right (only forward, note!). Then there are a couple of pieces that have no obvious equivalent in western chess. Lances sit on the outermost columns, and can only move forward, but can move any amount forward. Gold generals can move one space in any direction except the reverse diagonals. Silver generals can move on the backwards diagonals and any direction forward by one space. Thanks to Robert Johnson for teaching me shogi! I hope to play enough sometime to get a feel for how the game really works.

July 17, 2016 at 04:04AM

Today I Learned: 1) There's a variant of chess called Crazyhouse, in which you can place pieces captured from your opponents anywhere on the board except the last rank in lieu of moving a piece on your turn. That rule is also standard in the Japanese version of chess, shogi. Checking out Crazyhouse led me to the awesome wikipedia article on "List of chess variants". A couple of fun variants from there include absoption chess, in which pieces gain the movement abilities of pieces they capture; genesis chess, in which the board starts empty and you place your pieces as a move; synchronous chess, in which the players record their moves and execute them simultaneously, a la Diplomacy; and Beirut chess, in which each play secretly straps a "bomb" onto one piece, which can be detonated to take out all pieces in the surrounding squares (and can be used to kill kings). Turns out Erik Jue knows a lot of chess variants. 2) Protein translation can, rarely, initiate from codons other than the canonical AUG and its variants. Translation from other codons tops out around 2% of AUG translation initiation rate. 3) E-cigarettes smell kind of like cotton candy, at least at range.

July 16, 2016 at 03:32AM

Today I Learned: 1) Honey badgers are really good at getting out of enclosures, and they really enjoy the process of escape. They're extraordinarily clever about executing escapes, and will use tools, cooperation, and sheer grit to worm their way out of a surprising variety of prisons. See http://ift.tt/29EWS5S for details. 2) A QALY is a Quality Adjusted Life Year. It's a number used to calculate the value of life with different medical problems/conditions. By way of example, suppose you have a terminal illness that will kill you in a week. There are two mutually exclusive treatments. One will stave off the illness completely for X years, then will wear off and you will die. The other will completely cure the illness, but has the permanent side effect of giving you occasional painful stomach cramps for the rest of your life. Which do you take? If you're a doctor, which do you recommend? One standard way to evaluate this choice is to say that every year with stomach cramps is worth some fraction of a healthy year. By varying X above and asking a lot of people which treatment they would take, you can get a quantitative sense of how much people value not having stomach cramps. To figure out how much a treatment is worth, you multiply that value by the number of years you expect to get out of the treatment, giving you some number of year-equivalents of value for that treatment. This is what we call a QALY. Today I learned of a very similar, closely related concept called the Disability Adjusted Life Year (DALY). DALYs are very similar, in that they attempt to quantify the value of remaining life, except that they typically come from calculations of disease burdon rather than treatment gains. DALYs attempt to account for the loss of quality of life caused by the disease in addition to the straight-out loss of life it might cause. This fact brought to you by Erik Jue. 3) Either shelving units are super expensive or I've gotten far too used to cheap used furniture pricing....

July 15, 2016 at 12:26AM

Today I Learned: 1) Genbank files say right at the top how big their sequence is. (Genebank is a text format for storing sequence data. It's the primary format NCBI uses to display nucleotide sequence data. Example: http://ift.tt/29zmplA) 2) Gas stoves not only get to temperature faster than electric stoves -- their maximum temperature is also rather higher. Or maybe they just dump more heat into the environment. I guess I didn't actually measure the temperature of the flame. One second. Ok, measured between 100 and 250 °C at maximum flame, with the maximum at the center of the stove thingy, from the top. Can somebody repeat this for an electric stovetop? 3) There's some kind of insect out here that nibbles the edges off of thick tree leaves (not sure what the tree was -- something with a kind of oaky-looking leaf, but much waxier, and with smooth and peeling bark like a sycamore tree's) and leaves behind a lot of poop. I'll be keeping an eye on that particular tree to see if I can figure out what it is.

July 14, 2016 at 12:06AM

Today I Learned: 1) The DNA of a typical bacteria occupies ~1% of the cell's volume. That's interesting, to me, because I had it in my head that it was more like 10-50%, which confused me a bit because it doesn't seem like it would leave much room for all the other stuff going on in the cell. 2) A couple of astronomy-related numbers. First, the diameter of the sun is 109 times the diameter of the Earth (which makes its volume about a million times that of Earth). Second, Jupiter has more mass than *the rest of the planets in our solar system combined*. 3) More scales of things! The classical radius of an electron is about an attometer, or 10^-18. That also happens to be about the detection resolution limit on the LIGO, which is pretty ridiculous in my book.

July 13, 2016 at 02:56AM

Today I Learned: 1) Zinc finger proteins have really strong binding, based on some unpublished data I saw. Like, really, really strong binding. 2) One of the trees outside our house appears to be infested with some kind of Dictyostelium-like fungus -- its leaves have broken out in little stalks, spaced roughly an inch apart, tipped with little white heads. I've never seen anything quite like it. 3) ... a couple of benchmarks for visualizing the concentrations of solutions. 0.5 molar copper nitrate is packed with an average spacing of about 10-15 coppers' worth of distance (1.5 nanometers) between adjacent copper atoms (the rest being filled with water and nitrate). Hundred-micromolar primers are surprisingly not-dense, with an average distance of a micron or two between primers. I remember doing this calculation before and finding that they were much tighter than that, so if somebody could triple-check my math, I'd appreciate it.

July 11, 2016 at 01:13AM

Today I Learned: 1) When fingers and toes wrinkle in water, that's not osmosis. It's actually an active reduction in extremity volume caused by constriction of blood vessels in response to the sensation of water. If you cut the nerves to an extremity, it will no longer wrinkle in water. There's some evidence that this may be an adaptation to help grasp things better in water -- it turns out that although wrinkles do nothing noticably useful in air, it makes gripping objects underwater easier. Personally, I'd say it's just as likely that it's an adaptation to minimize heat loss by reducing blood flow to extremities. Or maybe it's both. OR MAYBE I SHOULDN'T BE SO TELEOLOGICAL. 2) Antibodies are big! Antibodies can come in a lot of sizes, actually, but a typical one is between 100 kD (100,000 Daltons) and several hundred kD. For a protein, that's quite large! Somehow I had it in my head that antibodies were tiny little proteins. Why did I think that? 3) Most (>75%) of the surface area in a 96-well plate is the well sides, not the bottom. Credit for this fact, and the second one, goes to Erik Jue.

July 10, 2016 at 04:08AM

Today I Learned: 1) ...how to keep a credit score high. Most important thing is to pay your bills on time, but it's also good to have multiple credit cards, keep multiple accounts open of different types, keep accounts open for a long time, and keep your credit draws well below your credit limit. None of which are things I do (except for paying bills on time), yet my credit score is pretty fine, so I'm not sure what's up with my credit score. 2) No, X-Gal will not dissolve in 5% DMSO in water. As much as I want it to, it just won't. 3) Speaking of X-Gal, it turns out that most plasmids used for blue-white screening... wait a second, let me back up a bit. Yesterday I explained what X-Gal is. In brief, it's a molecule that turns blue when the enzyme β-galactosidase is present. Why would you want such a molecule, though? Well, the answer is that it's useful for diagnosing certain failure modes when constructing a plasmid (a circular piece of DNA with up to a few genes). There are lots of ways to build a plasmid, but most of them go something like this: you start with a pre-existing plasmid that produces some sort of antibiotic resistance (called a "vector") and some kind of DNA you want to insert on the plasmid; you mix the two with some enzymes that will hopefully insert your DNA into the vector; you forcibly insert whatever you get into a bunch of bacteria and grow them up on agar plates at low density, along with an antibiotic that the plasmid confers resistance against; each bacteria that got a whole plasmid will survive the antibiotic purge and grow up into a visible colony of a whole bunch of bacteria; you can then pick that colony (as in, literally pick up with a pipette tip), grow it up in media, and harvest your new plasmid. One of the (many) ways plasmid construction can go wrong is that the original vector will sometimes just survive the construction process and get into a bacteria unmodified. Blue-white screening is a way of quickly identifying which bacteria have unmodified plasmids. To do blue-white screening, you start with a vector that expresses β－galactosidase, and you do your construction in a way that, if done correctly, will remove or destroy the β－galactosidase gene. Then, when you grow up your plasmid-transformed bacteria, you do it on plates with X-gal, so that the colonies with unmodified plasmid look blue. So, that's blue-white screening. Today I learned that most vectors for blue-white screening have β-galactosidase under a lac promoter, which is naturally repressed. To actually produce the β-galactosidase and turn colonies blue, you have to *also* add IPTG, a molecule that lifts lac repression. I tried making some stocks of mixed IPTG and X-Gal solution, then used that to make plates for blue-white screening. We'll see how well they work.

July 08, 2016 at 10:53PM

Today I Learned: Here are a couple of TILs I've been hoarding over the last couple of days while I waited for internet: 1) Solarpunk is a movement/genre that, broadly speaking, envisions a future where humanity is highly technologically advanced, highly re-integrated into nature, highly decentralized, and values individual and communal human life. Unlike cyberpunk, it is a distinctly utopian form of futurism; unlike steampunk, it is a radically forward-looking movement. It also has one of the most beautiful google image searches I've seen in quite some time (http://ift.tt/29tL2hL, for the lazy). If you ever want to know my best-case vision of the future, it looks decidedly like solarpunk. 2) ...how Pokemon Go works. Pokemon Go is an augmented-reality Pokemon game, played on any mobile phone. The game consists of static encounters in the world, some randomly generated (most wild pokemon, which appear in locations relevant to their type, i.e., water-type pokemon might appear only near water sources), some fixed (real-world landmarks corresponding to in-game places of interest, like gyms). When you come across a wild pokemon, you can attempt to catch it using a simple finger-swipe-based minigame. When you come across another trainer, you can battle your respective pokemon with another swiping-based minigame. There's an additional social layer to the game, in that you can join teams of trainers and battle for control of gyms, which again, are located near real-world landmarks. You can capture gyms, then leave pokemon to guard the gym against other teams. If they battle and defeat your pokemon sufficiently often, you will lose control of the gym. I'm not sure how many teams there are, or how big they are, but I'm curious to see how well the Pokemon Go social dynamic turns out. 3) X-gal is a molecule similar in structure to lactose, which is a dimer made of a galactose molecule and a glucose molecule. The enzyme β-galactosidase, which normally cleaves lactose, will also cleave X-gal, but instead of releasing a galactose and a glucose, X-gal breaks down into a galactose and a blue-colored indole. This makes X-gal a sensor for β-galactosidase -- if you add X-gal to something and it turns blue, it has β－galactosidase (or something else that can break down lactose-like molecules). Today I learned that X-gal is a little annoying to work with, in that a) it is typically dissolved in DMF, a somewhat toxic solvent, and b) you need a fair amount of it to make agar plates with X-gal in them at working concentration.

July 07, 2016 at 10:43PM

Today I Learned: Here are a couple of TILs I've been hoarding over the last couple of days while I waited for internet: 1) US government bond interest rates are some of the best in the world, despite being solidly less than 2%. Many countries' government bonds actually pay *negative* interest -- they're basically a way to store your money very, very safely, for a small fee. 2) There are two "kinds" of ultraviolet (UV) light, called UVA and UVB. UVA is higher wavelength, closest to visible light. UVB is lower wavelength, down to 290 nanometers. There are difference in how the two kinds of UV tend to be absorbed in the body, but what interests me is their transmission through glass. UVB is almost completely blocked by basically any kind of glass. UVA will penetrate most basic types of glass, but some kinds (laminated, in particular) are UVA-proof. Why do I care? Well, the old way of visualizing DNA in a gel is to stain it with a dye like ethidium bromide and blast it with a ton of UV light, which causes the stain to light up in the visible spectrum. I've been setting up to do this, but the problem is that it produces enough UV to be quite carcinogenic (not to mention damaging on the eyes). I've currently got the UV source in a glassed-in box, but the UV source can generate either 302 nm (UVB) or 365 nm (UVA) light. Fortunately for me, ethidium bromide absorbs at 302 nm, so as long as nobody ever switches on the other setting, I'll be fine. 3) You can make a really compelling supersaturated solution demonstration out of sodium acetate (the salt you get when you combine baking soda and vinegar). You can melt a bunch of it and cool it to room temperature, and if you do it right it should stay liquid until you give it some solid crystals to nucleate around. You can try dusting your finger with some sodium acetate crystals and sticking it in the solution -- it will quickly coat your hand in what looks kind of like ice, but is actually quite warm (the freezing reaction, like all freezing reactions, releases heat). This stuff is colloquially called "hot ice".

July 05, 2016 at 03:44AM

Today I learned the sound of a shattering ceramic mug. My beautiful dragon mug now lies in about sixty pieces big enough to be visible (yeah, I counted). Were I a Buddhist, this is where the rubber would meet the road -- this is a perfect example of when it's important to give up your desires. But I guess I'm not a Buddhist, because this is not a loss I will not accept. There is glue which will fix this mug, and I will find it, and I will get it back in drinking condition. This is a sobering reminder of how quickly pretty much anything can fall apart. All it took was a second of carelessness. At least all it cost me was, at worst, a mug and some time. Stay vigilant, and don't take those around you for granted. I think I'm going to call it a night. Oh, and for those who are still curious, it sounds a *lot* like a sturdy plastic cup hitting the ground (like a cafeteria cup, not like a solo cup), but with some additional high-frequency power and less echo.

July 04, 2016 at 04:52AM

Today I Learned: 1) ...lots of game rules. Learned about Resistance (a variation on Mafia where nobody dies and there's no moderator), Princes of Florence (a hybrid bidding/building placement/action-based, victory-point-scoring German board game about burning money to show off artists and scientists as ornately as possible), and chinese chess (which is surprisingly similar to chess, but with generally weaker pieces, more movement restrictions, and a pretty wacky rule that kings can jump all the way across the board to capture the other king if they have straight line-of-sight). New game rules mostly courtesy of Robert Johnson. 2) A two-inch mantis can't really get through a junebug's armor. Not that it won't try, mind you.... 3) ...a couple of facts about Cas9-based activators in Eukaryotic systems. There was a nice little paper in Nature Methods at the beginning of the year that I finally actually read through today, directly comparing eight different activator variations in a couple of human cell lines(http://ift.tt/29d3r3U, behind a paywall). Three of them worked particularly effectively and robustly: * VPR is a modified version of the standard Cas9-VP64 activator fusion protein (literally just a non-cutting Cas9 fused to the transcriptional activator VP64), with two other activators (I think) tacked onto the VP64 part of the protein. * SAM is another VP64 variant where the guide RNA is modified with some target sequences for an RNA-targeting/gene-activating fusion protein. The combined complex of VP64 and the other gRNA-targeting activator gives SAM a little more oomph, and it looks like SAM was the most robust activator studied in the paper. * SunTag is a system with an aptamer-modified version of dCas9 and a fusion of some antibody fragments (scFV) and VP64. The aptamer fused onto dCas9 consists of 24 repeats of a short peptide sequence that the antibody fragments bind to strongly, bringing VP64 with them, so it's effectively a dCas9 molecule with 24 VP64s attached. Aside from some straightforward characterization stuff, there were a couple of specific interesting points from the paper. Firstly, all of the activators produced the highest fold-change in genes with naturally-low expression, suggesting that there's an inherent limit to how much a gene can turn on, and that strong activation with some of these constructs might bump up against that limit. Secondly, through RNA-seq, the authors could not detect any off-target activation, so these activators are quite specific (which is a little surprising, but nice!). Thirdly, the authors tried making combinations of the three successful activators, but none worked any better than any activator alone. Just how to interpret that is still kind of a question mark.

July 03, 2016 at 05:50AM

Today I Learned: 1) LEGOs are surprisingly light. Very convenient if you're moving lots of them around. 2) Lots of plumbing features (toilets and sinks, in particular) have values with oval-shaped handles. When the long side of the oval is aligned with the pipe, it means that the valve is open. Andrey Shur taught me this trick, and then I used it a couple hours later to fix a toilet. Let's hear it for plumbing knowledge! 3) I was forced to improvise lunch from a non-optimal set of food choices today, which, luckily, made me try something I never would have tried that turned out much better than I hoped. In particular, today I learned that a hummus and pickled daikon radish is way tastier than it sounds. The combination of sweet and refreshing and crispy from the daikon meshes pretty well with the smooth savoriness of hummus.

July 02, 2016 at 01:59AM

Today I Learned: 1) Thermal cyclers, for those who don't work with thermal cyclers, are machines used in biology labs to heat up and cool down tiny tubes of liquid in a very controlled way on an automatic program. There used to be this problem with thermal cycling liquids that you typically have to ramp up the temperature to above 95 C more than once, which tends to evaporate your liquid mix. That might be fine, except that it could then condense on the top of the tube. This movement of liquid out of your sample could seriously concentrate your sample, which could easily make enzymatic reactions stop working as they got to salty or too overloaded. The old-school solution to this problem was to put a layer of mineral oil on top of your samples, which would prevent evaporation. Modern thermal cyclers instead cover the sample tubes with a metal lid heated to 105 C. That keeps the lids hot enough that water can't condense on them. There will still be some evaporation, but the air inside pretty quickly saturates with water vapor, and the effect is pretty small. Today I learned that some older heated-lid thermal cyclers will shut off their heated lid when the run is done. That matters because usually, when runs finish, the thermal cycler drops the samples down to 4 C (refrigerator temperature) to keep them cool and protected. With a heated lid, that isn't a problem. Without a heated lid, that's kind of a problem, because the block where you put your tubes starts to condense water out of the air. If you leave that running overnight, it can pool up quite a bit of water. 2) How window air conditioners are installed! More or less. At least, a little. In particular, I learned about the kinds of insulation material you can put around the outside of an air conditioner, and about some strategies for filling in parts of the window that aren't covered by the air conditioner. 3) A few pricing things: Doors can be bought surprisingly cheaply (~$50) and surprisingly expensively (~$1,000). 3D printing material goes for order-of-magnitude $25. Glass gets really expensive as you start trying to fill things like windows. Desktop computers are surprisingly not *that* much cheaper than the last time I checked, which was a couple of years ago. A $200 desktop at Fry's Electronics gives you an Intel i3 processor, 4 GB of RAM, a terabyte hard drive, and not much else. I guess that's really cheap, but I wasn't all that impressed. Water coolers cost in the range of $90-$200. Not really a pricing thing, but a lot of screw packs, especially for drywall screws, come with free drill bits. How nice!