January 30, 2016 at 10:59PM

Today I Learned: 1) One of the developers of Titanfall spent his two-week vacation porting the game to Oculus rift, and the dev team tried it out. The conclusion? Titanfall was a terrible game for Oculus rift. Why? In Titanfall, you spent a lot of time moving very quickly and flying through the air, which is *terrifying* in Oculus rift. 2) Aphaenogaster cockerelli (the species of my new ants!) apparently like insects and seeds, but not sweet stuff like honey, fruit, or sugar water. 3) A few facts about genome sequencing: market price for a full-exome sequencing run, with analysis and interpretation, for a medical institution, is around $4,000; the most common way to get DNA for sequencing (again, in a medical context) is to extract it from blood. This seems a little crazy to me, since the vast majority of cells in blood have no DNA in them...; when DNA is purified and shipped before sequencing, rather than shipped in blood form, it is much more difficult to predict copy number variations, for reasons that are not entirely understood. Thanks Jeanne Morin-Leisk!

January 30, 2016 at 02:34AM

Today I Learned: 1) Don't empty bottles of LB-agar into a sink. You'll end up with lots of annoying chunks that are hard to clean up. Just dump it out in the trash can to begin with. It'll smell really funky, but it's not actually that messy and it ought to be totally sterile. 2) The Plank length is the fundamental unit of length in nature, and is thought to be effectively the smallest size a thing can be. ...or that's what I thought. Apparently that's a common misconception. Today I learned that there is no such thing, and it makes no real sense to talk about a "smallest size of thing" (the reason is relativity -- if you made something of the "smallest possible size", there would always be a reference frame you could come up with where that thing was smaller due to length contraction). The Plank length IS a thing, just not THAT thing. It's roughly the distance at which you have to stop ignoring quantum gravity, which is a concept I don't really understand. According to Wikipedia, which I may or may not trust on this matter, it's also roughly the smallest *measurable* distance. That's not an experimental limitation -- it's just that below that scale, traditional notions of distance aren't very useful because stuff is smeared out too much. Also according to Wiki (and I trust this fact more), a 0.1 millimeter dot (about the smallest speck you can see) is as much bigger than the Plank length as the universe is bigger than a 0.1 millimeter dot. That's a really small unit of measurement! The other cool thing about the Plank length is that it's defined entirely in terms of fundamental physical constants, so it doesn't depend on your choice of units (just your choice of universe). 3) 2 AA batteries (3 volts) is sufficient to split water with a graphite cathode and a graphite electrode. 1 AA battery (1.5 volts) is not. Slightly relatedly, volt meters apparently tend to use a special 12-volt battery that looks like a AA battery squashed down to the length of a AAA battery.

January 29, 2016 at 05:09AM

Today I Learned: 1) There is a website called "willyoupressthebutton.com" which gives you a hypothetical button to press, or not press. If you press the hypothetical button, you get some really good thing, stated by the site, but it is accompanied by something really bad, also listed. For example, one button I ran across would let you visit any place in the world for a week, but the whole time you would be there, a large spider would be nesting in your hair. Do you press the button? Another one: You will become immortal and live forever, but everyone will hate you. Do you press the button? Warning -- the site is lightweight but add-heavy. If you find this sort of site annoying, do not go there. Warning -- this site has an extremely fast action loop and high addiction potential. If you do not have time to spare, do not go to there. 2) The Hanon exercises, a staple of piano education, were written in the mid 19th century by, well, Hanon, who was FRENCH. For more than a decade, I have assumed that Hanon was Italian, or perhaps Austrian or German. Not so, not so. 3) One of the biggest rate-limiting steps on game developers deploying patches is vendor quality assurance, or QA. Any developer has to put new code through rigorous tests before deploying it, or it is pretty much guaranteed to have bugs, which in the case of games, can be, well, game-breaking. Testing takes time and resources, which means it can take a while for patches to get from developers' computers to players' machines. Today I learned that even after a patch has been tested in-house, there is an additional layer of testing performed by the QA departments of the *platform* on which the software was released (i.e., Microsoft for an Xbox game, Apple for an iTunes app, etc). The vendor QA team makes sure that the new patch isn't malicious, doesn't break anything obviously, and otherwise meets the quality standards of the platform. This additional QA testing takes between 7 and 14 days for a patch for a typical AAA console video game, which can be a big chunk of the time required to churn out patches. Having a layer of vendor QA also incentivizes developers to spend longer testing and verifying their patch, because if anything goes wrong, it will remain live and unfixed for a MINIMUM of a week before any counterpatch can possibly be deployed. This and lots of other interesting tidbits on engineering the guts of video games from this tech talk at Rackspace by Jon Shiring, one of the engineers from Respawn Entertainment: https://www.youtube.com/watch?v=ayF8e8q_aA8

January 28, 2016 at 04:00AM

Today I Learned: 1) Polymerases have EXTREMELY high affinity for DNA. As a consequence, the vast majority of polymerases in any given cell are bound to DNA at any one time. This applies both to DNA polymerases, which replicate your genome, and RNA polymerases, which transcribe your genes into RNA. 2) High molecular weight DNA (i.e., genomes) is quite viscous. Check out any youtube video of at-home DNA extraction to see for yourself... ...or better yet, try out a DNA extraction yourself! You basically just need some fruit, high-percentage ethanol (>=70%), salt, and detergent. Try it out here: http://ift.tt/1QuhdLv. Note that this procedure isn't sensitive to much except, to some degree, the makeup of the lysis buffer. If you don't have 50 mL conicals at home (and who does?), just use a glass. If you don't have a coffee filter, use a paper towel or colander or anything else that will strain out the big chunks. 3) ...what the (valid English) sentence "Buffalo buffalo Buffalo buffalo buffalo buffalo Buffalo buffalo" means. It uses three versions of the word "buffalo": the animal otherwise known as a bison; the city in New York; and the archaic use meaning roughly "to bully" (or possibly to bully roughly?). The 7x buffalo sentence translates to, it's "New York bison that New York bison bully, bully New York bison." Thanks to Robert Johnson for explaining this one to me, and also for pointing out that you can take it to 11x buffalo quite easily -- "New York bison that New York bison bully, bully New York bison that New York bison bully".

January 27, 2016 at 01:00AM

Today I Learned: 1) Neil DeGrasse Tyson has a nephew, Stephen Tyson, who raps. Look him up and you'll run across a rather amusing, if slightly alarming, recent story. 2) The E. coli-based cell-free extract system we use in lab, TX-TL, is MUCH better at producing GFP than similar commercial kits. TX-TL is essentially cell innards minus the membrane, while commercial kits are built from the ground up with known components. Also, TX-TL is an order of magnitude or two cheaper than commercial systems, which is nice. 3) "Tort reform" is any change to the legal system that limits the ability of people to suit. In the US, tort reform almost always refers to the practice of capping payouts for medical malpractice payouts. The goal of tort reform (in the context of health care) is to reduce the amount of suits brought against doctors for malpractice and to reduce the payouts associated with those suits. This, in turn, should lower medical malpractice insurance, lowering the cost of healthcare. Does it work? Almost unambiguously no. Malpractice insurance only accounts for a couple of percent of US health care spending anyway, so even a massive reduction in malpractice insurance costs wouldn't do much to bring down overall costs. There's also no evidence that tort reform has lowered healthcare costs *at all* in the states where it has been implemented. Florida is an interesting case study -- when tort reform was enacted in Florida, the primary effect was that insurance companies didn't have to pay out as much money on lost litigation, and they didn't lower their prices, so their profits went up by a factor of 40 (not a typo).

January 26, 2016 at 02:24AM

Today I Learned: 1) ...how to reserve meeting rooms at Caltech. 2) Marvin Minsky, one of the founders of AI research, was still alive until yesterday! ... He just died from a cerebral hemorrhage. I also learned, from Minsky's wikipedia, of a nice little hacker koan from the Jargon File featuring Minsky, which I will reproduce here. (Incidentally, if you haven't checked out the Jargon File, you should) In the days when Sussman was a novice, Minsky once came to him as he sat hacking at the PDP-6. "What are you doing?" asked Minsky. "I am training a randomly wired neural net to play Tic-tac-toe," Sussman replied. "Why is the net wired randomly?" asked Minsky. "I do not want it to have any preconceptions of how to play," Sussman said. Minsky then shut his eyes. "Why do you close your eyes?" Sussman asked his teacher. "So that the room will be empty." At that moment, Sussman was enlightened. 3) There are a lot of ways in which people believe they can tell the sex of an unborn baby. They are all wrong except for a) ultrasound or other imaging techniques that can actually look, c) genetic tests, or d) only using embryos pre-screened for one sex. There is also a very slight but statistically significant correlation between extreme, debilitating morning sickness and having a daughter rather than a son. Many, many other techniques have been systematically tested and shown to be no more accurate than chance, including measuring the belly of the mother, measuring heartate of the baby, intensity of morning sickness (again, only slightly and only for the strongest morning sickness!), and a weird test involving Drano.

January 25, 2016 at 03:50AM

Today I Learned: 1) ...how to steam-sterilize soil! I wouldn't use this as a substitute for autoclaving for truly sensitive tasks, but apparently steaming (preferred) or baking soil to about 180 F sterilizes it pretty well. It's also supposedly important to not heat it much over 180 F, or it will "create toxins". 2) I learned a few things about my handwriting today. Erik Jue and I compared how we write our letters today, and we discovered that I've heavily optimized my handwriting for efficiency of movement, with a couple of exceptions. I also have a strong tendancy to start letters from the bottom and pushing up rather than from the top pulling down. These two observations probably explain a lot of why my handwriting is as unreadable as it is. The two of us each came up with a list of stroke orders to change. For instance, I have an egregiously wasteful 'r', and my 'k' is a little less efficient and innaccurate than the more standard protocol. I've also decided to re-work my 'g's for better control, at the expense of some speed. We also discovered a slightly better way to write 'j' than either of us were using -- we both start at the top of the main part of the letter, stroke down, then go back up to dot the j. It seems just better to dot first, then add the main part underneath, saving yourself potentially a full character-height of travel time. Thanks, obviously, to Erik Jue. 3) I've looked up what tensors are many, many times, and never quite figured out what they are. There's something a bit sideways about how they're defined that somehow makes them slippery to my mind. For instance, almost every source on tensors begins by stating that they are generalizations of scalars, vectors, and matrices. This is only slightly helpful to me. There's more than one way one might imagine generalizing a matrix, for instance. I think I understand at least in part what a tensor is now. In programming terms, a tensor of rank N is a function that takes N index arguments and returns a number. For example, a matrix is a rank 2 tensor, which takes two indexes (a row number and a column number) and returns a number (the value of that row and column). A scalar (otherwise generally known as a number) is a rank 0 tensor, which takes zero indexes and returns a number. Now that I write it down, it seems so simple, but it took me a long time to figure out that function signature. Now, there's a bit more to tensors, as they also have to satisfy certain properties about invariance under changes of coordinate bases, which leads somehow to a more intrinsic definition, but I think this is a pretty good first-order approximation. Thanks also to Erik Jue!

January 24, 2016 at 02:58AM

Today I Learned: 1) Chickadees are pretty crazy little metabolic machines. I mean, I knew that songbirds tend to be super-energetic and eat a lot to survive, but today I learned a few numbers to put it in perspective. To give you an idea of what a chickadee has to do to survive the winter, chickadees eat seeds to put on a visible layer of fat *every day* amounting to about 10% of their body weight, just about all of which is burned overnight to stay alive. 2) Apparently I should be reading the Galactic Center books by Gregory Benford. Can anybody confirm or deny? Also, I'm quite sure I've heard of Benford before in some other context -- where might I have seen him? 3) Bowhead whales are thought to live to be over 200 years old. How, you might ask, could we possibly know this? It's not like anyone's been tracking individual whales for 200 years, and they don't have anything like tree rings that you can tell age from. Turns out we know (or at least very strongly suspect) the longevity of bowhead whales because a specimen was found recently with a 200-year-old crossbow bolt stuck deep in its muscles.

January 23, 2016 at 12:12AM

Today I Learned: 1) Babies are weird. Like, more than I knew. Item 1: Young babies perceive the world through a much clearer eyeball than we adults, and are much more sensitive to light. Item 2: Babies hear echos, always, generated by the difference in timings between each ear receiving a new sound. Your ears pick up the same echo, but somewhere in your auditory pipeline you learn how to process these echos into a single coherent sound (with directionality). Don't ask how we know this. I'm just trusting Radiolab on this one. Item 3: Babies under brain scans sometimes, apparently at random, will respond to stimulus as though they experienced a stimulus of a different modality. For example, a baby might be shown a circle but its brain will light up as though it heard a sound (and presumably it *does* hear a sound). 2) You know how you can add the voltages of batteries together by wiring them up in series? Turns out you can also *subtract* voltage from a battery by wiring it to another battery in series, but backward. 3) There is probably almost never a reason to run PCRs for more than 25 cycles. A quick back-of-the-envelope calculation shows that you run out of primers sometime around then under ideal conditions (though I've read from some people who studied PCR quite extensively in the 90s that primers are never limiting in PCR....), and everything you amplify after that is off-target messy stuff.

January 22, 2016 at 06:36AM

Today I Learned: 1) The Mormon chruch has a board of living saints, who publish new scripture at least yearly, I think more often. It's kind of interesting having a religion whose fundamental writings include things being written *today*, for issues of today. I also wonder how modern Mormon scripture differs from, say, non-scriptural sermons of other Protestant faiths. 2) Laboratory refrigerators are expensive! Under-desk models run in the thousands of dollars. I'm sure they're better than the average refrigerator, but... an order of magnitude better? I don't think so. 3) Racemases are proteins which... actually, before I dive into this one, let me say a couple words about enantiomerism, or left- vs. right-handedness in molecules. If you know what an enantiomer is, feel free to skip the next two paragraphs. Two chemicals with the same component parts don't necessarily act at all the same. The arrangement of atoms in a molecule, as much as the types and numbers of those atoms, determines the properties of that molecule. This is generally pretty easy to recognize -- six carbon atoms linked in a straight chain aren't going to act the same as six carbon atoms arranged in a ring, for instance. A more subtle way molecules with the same atoms can differ is by their *handedness*. Consider a pair of gloves. Each glove is, in a way, substantively identical to the other. They both have four finger pockets and a thumb pocket arranged in roughly the same way... but one is a mirror image of the other. There's no way you can rotate a left-handed glove around to make it a right-handed glove. Some molecules can be the same way -- identical except that one is a mirror image of the other. When this is the case, we say that the left-handed version is one "enantiomer" (or "stereoisomer"), and the right handed version is another "enantiomer". Two different enantiomers of the same molecule have identical physical properties except that a) they interact with polarized light differently, and b) they interact differently depending on the enantiomer they interact with. As a metaphor, left-handed and right-handed gloves don't really have any different physical properties, but one only interacts properly with a left and and the other only interacts properly with the right hand. Enantiomers are really important any time biology gets involved with chemistry (or vice versa) because biological molecules generally only come in one enantiomer, or if they come in more than one, the different enantiomers typically have totally different biological properties and activities. The most famous example is thalidomide -- one stereoisomer is a very effective anti-nausea drug, but the other is a potent teratogen (it mutates babies). When thalidomide was deployed widely in hospitals in the 1950s, the drug was produced in both enantiomers, leading to thousands of cases of deadly mutations in newborns. Thus, chemists in a lot of contexts really, really care about making chemicals in one isomer over the other. Unfortunately, in general, it's really hard to selectively make or even isolate one enantiomer of a compound, because both isomers have identical physical properties! Pretty much all techniques for stereoselectivity involve using isomerically pure substances from natural sources. Back to racemases. Racemases are a class of enzyme that switch enantiomeric centers from one handedness to the other. They generally do so non-specifically, meaning that they will convert from either handedness to the other one (though usually one more efficiently than the other). To the cell, this is just another form of biochemial synthesis available to make new compounds. To a chemist, this is a wonder protein. "But wait," you might ask, "why would I want a molecule that turns pure solutions of an enantiomer into a mix of both enantiomers?" Good question. With some neat tricks, you can actually use racemases to do exactly the opposite, is the answer. Imagine you have a 50/50 mixture of two enantiomers of a molecule A, call them A_L and A_R. You want to convert the whole thing into A_L. Here's what you do: Add some racemase to the mixture. In the same pot, run a one-way reaction that converts A_L into something else (call it B_L) that's chemically inert in that setup (for example, by adding a protective group that blocks access by the racemase). Now B_L acts as a sink of A_L. As A_L is depleted, the racemase will start converting A_R into A_L to bring the two enantiomers into equilibrium. B_L can still pull A_L, though, so this set of reactions will suck all of A_R and A_L both into B_L. Once everything has been converted to B_L, you remove the racemase and run another reaction to turn B_L back into A_L (by, for example, removing the protecting group you added before). Amazingly, the racemase has helped you convert everything into one enantiomer! Thanks to Anders Knight for teaching me about these amazing proteins!

January 21, 2016 at 05:28AM

Today I Learned: Today's TIL comes from a lecture by David Baltimore, one of the rather more prestigious names I've had the privilege of listening to. The lecture was on human gene editing, and unfortunately for me it was a lecture for general scientists (mostly Caltech undergrads), so I knew most of it. There were still some nice tidbits, though. 1) About 15 base pairs of sequence is sufficient to be unique in the human genome. Either that's a very rounded number, or that's surprisingly low -- there are something like 6 billion bases in the human genome, so you would expect log_2(6 billion) ~ 32.5 bits ~ 16 bases to be unique. 2) Zinc finger proteins (ZFs) and TALEs are both classes of protein that recognize and bind to specific DNA sequences, and both can be modified to cut, much like Cas9. Both proteins are highly modular, with each segment binding to another little bit of DNA. ZFs and TALEs can be redesigned to target different, arbitrary sequences by rearranging, adding, and removing segments, but the process is rather arduous. Anyway, in this lecture, I learned that ZFs a) were discovered first, and b) have segments that target several bases at once, so the library of ZF subunits is on the order of dozens of variants, and c) are proprietary in their recombinant form, or at least were up until recently?. TALEs, on the other hand, a) were discovered after ZFs, b) are native to plants, c) have subunits that recognize individual nucleotides, so there are only somewhere around 4 variants, and d) are in the public domain. Basically, TALEs are strictly better for a bioengineer. 3) "Most" Mendelian diseases (that is, diseases caused by a single DNA mutation) are dominant. That surprised me a lot, and I'm actually a little suspicious of this -- how could gain-of-function mutations with negative side-effects be more common than loss-of-function mutations with negative side effects? In general, breaking stuff is much easier to do by mutation than making new stuff, even bad stuff.

January 20, 2016 at 03:05AM

Aw man, just learned another TIL-worthy fact. When worker ants of the genus Pogonomyrmex encounter each other in the field, they will often skirmish briefly before separating. However, Pogonomyrmex mark a roughly circular region around their nests, designating a sort of "core" area. If a Pogonomyrmex in its core area encounters another Pogonomyrmex, it will very aggressively fight and recruit other ants. Usually, the foreign ant will assume a submissive posture, and then the home-turf ants will drag, carry, or haul it out of their core area and release it. How civilized! I have no idea what happens if two Pogonomyrmex colonies are moved so that their core areas overlap. I assume some sort of thermonuclear war erupts.

January 20, 2016 at 02:48AM

Today I Learned: 1) "Octopuses", "octopi", and "octopodes*" are all accepted pluralizations of "octopus" in standard English. Source: http://ift.tt/1PgiiJU * Pronounced "oct-AH-pa-dees", like an Achaean hero, not "OCT-oh-PODES", like a marvel supervillan or a bunch of monsters with eight feet. 2) Lyophilization, which is just scientist-speak for freeze-drying (btw, TIL that "lyophilization" is just scientist-speak for freeze-drying) is the process of removing the water from stuff while it's frozen. Which I guess is kind of obvious from the word "freeze-drying", but I learned it anyway. Lyophilization apparatuses (apparatopodes?) work by applying vacuum to a frozen sample while simultaneously refrigerating it to keep it cold. Under strong enough vacuum, water will sublimate off the sample and right into the vacuum hose, drying the sample. I also learned why you would freeze dry something rather than heat-dry, as we do with speed-vacs fairly frequently in the lab. The trouble with speed-vacs and other heat-drying methods is that, well, they heat the thing you're drying, which can be quite destructive if you're trying to dry, say, a fragile protein. Lyophilization is the gentler way to dry stuff. 3) Optical coherence tomography (OCT) is a an imaging technique akin to ultrasound, except that it uses infrared light instead of high-frequency sound. Like ultrasound, OCT gives a live, cross-sectional or 3D image of the target tissue. The inherent faster frequency of infrared light gives OCT much higher theoretical resolution than ultrasound, and modern setups can image down to about 1 micrometer resolution, which means you can just barely see single cells -- the images look almost like histological images, except that they're LIVE and IN VIVO. Unfortunately, OCT only works to a depth of about 2-3 millimeters in a human body, but that's good enough to get a single-cell resolution cross-section of a retina, in the eye, non-invasively. Welcome to the future, everyone.

January 19, 2016 at 04:38AM

January 18, 2016 at 01:23AM

Today I Learned: 1) ...so, yesterday I made the claim that the nuclear pore complex was the largest atomically-precise biological structure known to man. Turns out I was wrong. Very wrong. Mostly. The nuclear pore complex is about 120 nanometers in diameter, with a hole in the middle between 5 and 10 nanometers across (for scale, about 8,000 nuclear pores would fit across a typical period, and if a nuclear pore complex were scaled to the size of an american football, the actual channel would be a hole about a half-inch to an inch in diameter). That's pretty big. ...but not as big as titin. Titin is a protein used as a spring in muscles to confer their elasticity. It's a long, long chain that blobs up on itself; when stretched, the bonds holding the blobbing together break and it stretches into a long string. When tension is released, the protein folds itself back up into a blob, pulling the ends back in. Also, titin is absolutely freaking massive. Most proteins are a few hundred amino acids long. Titin is about *30,000* amino acids. Most proteins have mass measured in kilodaltons, or often tens of kilodaltons. Titin is just under *4 megadaltons*. Wikipedia helpfully informs me that the empirical chemical formula for titin is C169723°H270464°N45688°O52243°S912 (°s for clarity). Titin also has the longest technical name of any protein (for obvious reasons). From wiki: "The full chemical name of the human canonical form of titin, which starts methionyl... and ends ...isoleucine, contains 189,819 letters and is sometimes stated to be the longest word in the English language, or any language." That said, I'm not sure if you can call titin atomically precise. It's kind of floppy, and in fact flops around as part of its normal operation. Still, it has a more-or-less well-defined structure, and it's a heck of a lot bigger than the nuclear pore complex. 2) The US spends billions of dollars annually on fossil fuel energy. This worried me until I realized that a) it's only single-digit billions of dollars, which, while enough to show up as a blip in the US budget, only shows up as a blip in the US budget, and b) the US spends *more* billions of dollars annually on renewable energy. I thought for bit (hoped, really) that subsidies might explain why coal and natural gas are so much cheaper than the alternatives (though, that said, coal isn't as much cheaper as I remember it being), but then I crunched the numbers and realized that single-digit billions of dollars of subsidies annually, if they translated directly into customer savings, would still only provide a couple percent drop in price of coal energy. 3) Semi-rational protein design is way, way better than I thought it was. David Baker's lab at the University of Washington (one of the creaters of Rosetta and FoldIt@Home) have created a shocking number of proteins from scratch including: "Ideal" (regular, symmetric) globular proteins; protein helices held together with pair-specific hydrogen bonds, much like DNA but with three or more helices; proteins that form bundled fibers in various programmed configurations; small peptides that specifically bind to conserved flu coat protein motifs (and soon many other viruses)*; proteins that self-assemble into 2D space-filling lattices; sets of proteins that self-assemble into regular isocahedrons (d20s) of different, programmed sizes; and proteins that, when transcribed, can package their own RNA! Rosetta, the software that designs these proteins, starts with an empty tiny chain of amino acids, then iteratively tries adding different random amino acids, simulating the new chains' folding, and picking the ones that have the structure closest to the target structure. I'm guessing this takes quite a while to run, but it appears to be able to design proteins (and predict their structure) very, very accurately. One quirk of the software's design is that the proteins it makes tend to be extremely stable, to the point where some are essentially unable to be unfolded using standard techniques. * Apparently if you squirt these into ferrets and mice as an aerosol, it protects them against flu infection.

January 17, 2016 at 01:29AM

Today I Learned: 1) ...one simple technique for detecting manipulation of JPEGs called error level analysis (ELA). ELA is based on the fact that JPEGs (and JPGs) are compressed in a lossy way -- every time you resave a JPEG, it loses some information. To perform ELA, you intentionally resave the JPEG at a relatively low quality level, then compare the result to the original image. You can't usually see the difference by eye, but if you essentially subtract off the lossily compressed image, you can get a sense of how much compression the original image already had gone through -- if the ELA image comes out dark, it means the image was already quite compressed, so resaving it didn't do much to change it; if the ELA comes out bright, then the image changed a lot when it was compressed, so the difference was large. In any case, in a typical image, an ELA is more or less even (though usually salt-and-peppery) across the image (with some exceptions like patches of solid color, which compress very cleanly and so always come out black in the ELA). If part of the image was photoshopped, that part will usually have a different level of compression than the rest of the image, which can be seen in the ELA as a patch of wildly different intesity from the original image. How can you fool an ELA? I imagine you could just resave a JPEG a bunch of times to bring everything down to a low quality level. So now you know to be a bit more suspicious of low-quality JPEGs.... Thanks to Chris Lennox on this one! 2) DNA origami* is one of the coolest techniques on the planet, and one of the things I'm keeping my eye on to potentially radically change medicine, but it has a glaring flaw -- DNA origami structures really don't do well in physiological conditions. There are two main reasons DNA origami breaks down in vivo. The most-cited problem (the one I knew about) is that DNAses** tend to chew them up pretty rapidly; the other main problem, which I learned today, is that living bodies have much lower magnesium concentrations than the folding buffers usually used to make DNA origami. The magnesium ions are important for making the DNA "sticky"; at lower concentrations, the origami gets floppy and falls apart. Today I learned a technique for fixing the second problem and at least somewhat mitigating the first. The key idea was to supply another source of positive charges in the form of oligolysine, which is a bunch of the amino acid lysine strung together (lysine is positively charged and realtively small and therefore unlikely to get too much in the way). If you supply oligolysine at the right concentration, it stabilizes the origami. You still have to *fold* the origami in a magnesium buffer, but then you can add oligolysine, move the origami to living tissue, and it will be fine. Oligolysine also protects modestly against DNAses, probably just by getting in the way. You can further protect origami against DNAses with polyethylene glycol, or PEG, which is usually used in biology as a "crowding agent", or a chemical that takes up a bunch of space to effectivley make other things act like they're at higher concentration than they are. PEG also seems to get in the way of DNAses, and can be incorporated into origami by chemically linking it to oligolysines before adding them to the origami. Adding PEG-K10 (PEG attached to a 10-lysine oligo) extends the half-life of (some particular) DNA origami in physiological conditions (i.e., in cells) from about 3 minutes to about a day. That's a good start. *I've written about DNA origami before, so I won't go into too much detail, but basically it's a technique for folding a very large circle of single-stranded DNA into a stable shape of your choosing. It's ridiculously easy, remarkeably robust, and shockingly close to being cheap (it's about $1,000 worth of DNA to make one). ** DNAse = an enzyme that breaks down rogue DNA. 3) Yeast cells spend part of their life cycle as a haploid* and part as a diploid**. A diploid yeast can undergo one of the several kinds of fission yeast can undergo, resulting in two daughter cells with one set of chromosomes each. Depending on which version of chromosome 3 each daughter gets, it will either be a MAT-A or a MAT-α cell. MAT-A and MAT-α are a bit like male and female, except that one doesn't have a significantly larger reproductive cell than the other. When, however, a MAT-A and a MAT-α yeast come together, however, they fuse, combine chromosomes, and become a single diploid yeast cell. * haploid = has one set of chromosomes, like a sperm or egg cell. ** diploid = has two sets of chromosomes, like almost any cell in an adult human. 4) Bonus: I heard a claim (from someone much smarter and more experienced than I) that the largest known "atomically precise" biological structure is the nuclear pore complex. I'm unable to find a side-by-side to-scale comparison of the nuclear pore complex and, say, the ribosome, but I saw one earlier today and I was very impressed by the nuclear pore complex.

January 16, 2016 at 01:57AM

Today I Learned: 1) A metric space* is *complete* if for every convergent series in the space, the thing to which that series converges is also in the space. Intuitively, this means that it doesn't have any holes. Example -- [0,1], the set of real numbers from zero to one, including both zero and one. An example of a set that is almost-but-not-quite a complete metric space is (0,1), which is all of the real numbers from zero to one but not including either. In (0,1), for example, you can consider the series 1/2, 1/4, 1/8, 1/16/,... which converges to zero. Since zero isn't in (0,1), (0,1) isn't a copmlete metric space. A non-complete set can be "completed" by adding one or more points, which are the "completion" of the set. * a metric space is a set of stuff on which you can define a measurement of distance with some reasonable properties like the triangle inequality. Thanks to Dawna Bagharian!!!!! for setting me down a path of wiki articles to this information! 2) The Living Computer Museum! It's a museum of old computers, from the early 1960s through the early 90s, and almost all of them are functional and interactive. So you can go in and mess around on old 70s and 80s computers, or punch out punch cards with an IBM card-punching machine from 1964. It was a ton of fun to play around in. A couple things that stood out from browsing all the computers: 1) Very few of them had arrow keys! This made sense, though, since very few of those computers had any notion of backspacing. Several literally printed out your commands via typewriter as you typed, so an arrow key would probably just be ridiculous. 2) There was a lot of variety in auxilury key placement on keyboards (shift, ctrl, alt, break, insert, delete, those kinds of things). Even stuff like number and punctuation placement differed surprisingly between different keybaords up through the 70s computers. 3) The ergonimics of old keyboards were really different. The type-feel was in some ways much more like a typewriter than, say, my current macbook. 4) Mice are really recent! Also, they've gotten a lot better since they were introduced. 3) ...a story about how Microsoft got into the operating system business. Rumor has it Microsoft originally made some little auxilury software for IBM, which they sent representatives to IBM to give a talk about. Whatever IBM people saw it liked it, and went from that talk to a different company that was making an OS for them. The IBM reps didn't like what they saw, so they went back to the Microsoft team and asked if they could sell IBM an OS. The Microsoft team said they didn't have any OS, but they knew a guy who was working on one and could probably convince him to finish it, and then they could sell it to IBM. And thus, DOS was born.

January 14, 2016 at 11:03PM

Today I Learned: 1) Let's talk pulse oxymetry! Pulse oxymetry is the measurement of blood oxygen content by light (it's those little finger-clamp things for blood O2 concentration measurement). Today I learned that pulse oxymetry works because oxygenated blood and unoxygenated blood have different absorption spectra. Oxygenated blood absorbs red light more efficiently, and deoxygenated blood absorbs IR better. By measuring the absorption of both and taking the ratio between the two, you can account for and remove most of the noise. ...and oh boy, there are a lot of sources of noise in pulse oxymetry. For starters, there are differences in background absorption from skin and bone between people; differences in finger thickness; differences in light source intensity and sensor sensitivity. All of these should be corrected by the ratio-of-absorptions measure. The bigger problem is that venous blood is often less oxygenated in patients (it's actually pretty close to arterial blood oxygenation in healthy adults, most of the time). When a patient, say, moves their finger, venous blood sloshes around and messes up the reading. There are algorithmic ways to deal with this noise, which I won't go into here -- suffice it to say that finding the arterial blood signal in the wash of different noises of pulse oxymetry is one of the big algorithmic challenges of modern medicine. 2) Some fun numbers to know -- all estimates, taken from a talk, check before using these for anything important. All numbers are annual, for US only: 30 million people hospitalized 3 million hospital-acquired infections 200,000-400,000 preventable deaths in hospitals. These are deaths in people who were successfully recovering from whatever they went to the hospital for, but ended up dying from something else, like misapplication or misdiagnosis of drugs (or hospital-acquired infection). ~800,000 deaths in hospitals 3) Baby eyes are pretty sensitive to oxygen pressure. Pulse oxymeters, see, are often used with babies in ICUs to make sure they're getting enough oxygen. When the sensors detect a drop in oxygen, whoever's attending will usually raise the O2 pressure in the baby's chamber. However, the increased pressure can damage or even disable baby eyes. That's worth the risk if the baby would otherwise, you know, die, but for most of the history of pulse oxymetry, more than 90% of alarms were false alarms, so a lot of babies have been unnecessarily blinded . That's one reason it's really important to keep the false positives low with pulse oxymeters (the biggest reason is so that you don't waste all of the nurses' time with false positives and potentially) 4) Bonus fact: Carbon monoxide (CO) poisoning has chronic effects as well as acute effects. CO is an acute toxin because it binds ridiculously strongly to hemoglobin, displacing oxygen in your blood and suffocating you if you're exposed to too much. Turns out even if you *aren't* exposed to enough to suffocate you, it will stick in your system, making you more vulnerable to future CO poisoning and possibly cause nice little symptoms like increased risk of heart disease, neurological damage, and severe flu-like symptoms. Today's facts brought to you by Masimo Corporation. Not, like, in any sort of corporate sponsorship way. Just that the president told me all these facts. In a talk.

January 14, 2016 at 03:14AM

Today I Learned: 1) Continuing with my SpaceX-related posts, I watched part of a video tour of SpaceX headquarters, circa 2011. There were a couple of interesting tidbits -- for one thing, they have a shop that checks the precision of their machined parts down to about a thousandth of an inch to make sure they're perfect. Little details like that matter when building a rocket. The precision-checking room is kept temperature controlled at 68F -- if it wasn't, the metal would expand or contract, making precision measurement impossible. Part of me wonders why the rocket would ever work if tolerances have to be that good... but maybe it's ok if everything expands an contracts together? Or perhaps the really important parts are kept temperature-controlled during launch? Another interesting tidbit -- SpaceX's conference rooms are glass-walled, but those walls are rigged with motion sensors to turn foggy when someone enters the room. Musk seemed quite pleased about those walls. 2) ...fingering for a couple of minor scales I'd never learned before (for piano). 3) In longitudinal studies (not the best, I know), there is no evidence that smoking marijuana is associated with higher mortality or any obvious disease. This is a little surprising, since smoking tobacco certainly is. Are cigarettes just ridiculously efficiently packed with toxins? Would a fresh tobacco leaf rolled up in paper be safe to smoke?

January 13, 2016 at 02:52AM

Today I Learned: 1) Mammograms are around 85% sensitive and about 90% selective. That is, if you have breast cancer, a mammogram will tell you that about 85% of the time; if you don't have breast cancer, a mammogram will tell you you *do* have breast cancer about 10% of the time. 2) There are stationary-phase-specific promoters, which activate downstream genes only when bacterial density is very high (specifically, when RpoS (RNA polymerase sigma) is expressed). 3) ...what I'm supposed to do at TA sessions....

January 12, 2016 at 03:56AM

Tody I Learned: 1) ...a few soldering tips. For instance, you can use cool, thin-nosed pliers to scrape away hot solder in connections you don't want. Also, solder apparently tends to flow from cold to hot. Also, clamps are really, really useful when soldering. 2) Mono-winged aircraft (small ones, at least) require fewer control surfaces than large ones. 3) There is a relationship between the stiffness of different sequences of DNA and the stability of histone complexes* that form around those strands. The stiffer the strand, the more likely it is to unravel from its attached histones. Also, apparently cytosines (C bases) are by far the most likely to make a sequence more flexible. * histones are globular proteins whose primary purpose is to get wrapped around by DNA. This sequesters the DNA so it can't do things like get read to make proteins.

January 11, 2016 at 02:50AM

Today I Learned: 1) Ants lay pheromone trails to guide them from nest to food and back. How do they tell which direction is which? Today I learned a pretty comprehensive list of orientation mechanisms. The big three are light from the sun, polarization of light from the sun, and landmarks. Also, more rarely, species use: * Chemical cues from the nest: specifically, CO2 gradients, as the nests are pretty CO2-rich * Watching other ants: ants carrying food are more likely to be moving towards the nest * Forks in the path: if an ant comes across a fork with two paths at acute angles to each other, it can pretty safely assume the other path leads home. 2) Bone is piezoelectric! There's some debate on how much bone is actually piezoelectric and how much the observed piezoelectricity is actually due to streaming current (more on that below), but there's definitely some piezoelectricness to bone. There's some thought that this contributes to the feedback mechanism that causes strained bones to become thicker -- strain causes an electric current around/in the bone, which signals nearby cells to deposit more bone, strengthening the bone. Another thing to note -- apparently helical molecules tend to be piezoelectric, especially if they contain regularly-spaced charged groups. Collagen, for instance, is a helical protein that was one of the first-known biological piezoelectric materials, and is what is thought to give bone its piezoelectricness. DNA is also piezoelectric, though it's not known whether cells use this property. 3) Streaming current is a weird kind of electric current caused by physically forcing ions through liquid. It happens when you force an electrolyte (which is just a liquid with charged ions, i.e. a salt) past a rigid, charged surface. For instance, imagine you put some water with sodium chloride (table salt) into a tube with a positively-charged inner surface. The tube will attract some negatively-charged chloride ions, which will more or less stick to the outside of the tube. This makes the rest of the liquid net *positively* charged. If you then force the liquid to move, say by flowing more salt water into one end, the chloride ions will stay stuck to the tube surface but the positively-charged sodium ions in solution will move down the tube, so you have a net movement of charge, so you've generated positive current (remember that positive current is defined as the relative movement of positive charge). Why would you care about streaming current? Well, if you're a material scientist studying bone piezoelectricity, you have to be aware that liquids streaming through bone could cause streaming current if the bone is charged (and collagen has a lot of positive charges), so some of the "piezoelectricity" you measure could just be current streaming thorugh all the pores of the bone. You can also measure streaming current over a surface to measure how charged it is -- the stronger the charge, the more streaming current. 4) Bonus: SpaceX is hiring. A lot. I didn't count all the job openings, but I estimate about 350 (based on counting out 50 jobs here and estimating how many of those would fill the scroll bar: http://ift.tt/1cZ6GHD). Among other things, they're looking for IT guys, programmers, managers, business developers, and a whole bunch of engineers of all kinds.

January 10, 2016 at 03:44AM

Today I Learned: 1) Solid-state rockets (that is, rockets with solid fuel instead of liquid fuel) have the disadvantage that you can't stop them once you've turned them on -- the fuel is just a block, so once it starts burning, there's not a ton you can do to make it burn differently. In contrast, liquid-fuel rockets work by injecting fuel and oxidizer into an engine, so you can always stop injecting fuel. The advantage to solid-state fuel rockets is that they have fewer moving parts, for largely the same reasons. I also vaguely recall that they're more efficient, but I don't have anything to back that up. I have a nasty feeling I've already TILed this at some point -- anyone remember me posting this or similar information before? 2) Remember that M9 minimal media I mentioned a while back? In case you don't, M9 is a broth used for culturing bacteria. It's "minimal" in the sense that it's built from the ground up with known, quantified substances (typical broths like LB are mostly broken-down bacteria or yeast, so nobody *really* knows what's in them). Minimal medias are great because they're consistent and we know what's going on in them, but since they're pretty bare-bones in terms of nutrient diversity, bacteria tend not to grow as well in them. Today I learned about a possible problem with minimal medias. Most, it turns out, don't include trace metals like iron, selenium, nickel, and so forth, which are critical for growth. Yet bacteria somehow grow in those media(s?). How? It turns out that tap water contains plenty of trace metals (iron, for instance, is required at about 1 uM (micromolar, or about a 10^18 atoms in a liter)) for growth, and tap water usually contains several times this amount. Importantly, this means that if you make minimal media out of distilled water (or worse, 18 MOhm water!), you might find your bacteria don't grow as well. For a few minutes, I thought this might be the root of some problems I had a while ago that seemed to go away when I changed media batches... but then I remembered that I used a yeast-extract-based broth, not a minimal media. Oh well. This information, and many more fascinating facts, can be found in Cell Biology By The Numbers, by Rob Philips. It's about $50 on Amazon, or free here: http://ift.tt/1AvhcfE. If you've ever wanted to have a better quantitative idea of what living stuff *is*, I recommend checking it out. It's broken up into quite nicely-skimmable chunks, so I'd recommend it if you need some little chunks of reading to fill time. 3) SpaceX is quite operational, and has dozens of contracts lined up, which will keep it operational and building new rockets for many years. Also, SpaceX is currently valued somewhere in the neighborhood of $12 billion.

January 09, 2016 at 12:16AM

Today I Learned: 1) The elements Uranium, Neptunium, and Plutonium are named after Uranus, Neptune, and Pluto, respectively (in order of discovery). In retrospect, this should have been pretty obvious. 2) The space race was really expensive! I'm used to thinking of NASA as a super-cheap thing fed off the dregs of the US budget (currently NASA uses about 3% of discretionary spending, or half a percent of total spending, of the US federal government). During the 60s, though, NASA's budget spiked as high as 5% of national spending! That's the equivalent of $200 billion in today's dollars, which, for scale, is about a third of the military's budget (which itself is roughly the size of our yearly defecit). That's... like... big enough to make a dent. I'm a little flabergasted that we actually cared about space exploration enough to put that much cash into it. These first two facts and much more I got from WaitButWhy's series on Elon Musk. I highly recommend it -- if you're already a fan of Elon Musk, this will make you appreciate him even more; if you're not a fan of Elon Musk, this may come off as preachy w.r.t Musk himself, but there's a ton of cool information that comes along with it. Link here: http://ift.tt/1cbEGBL 3) There is a species of (almost) herbivorous spider. It's a jumping spider called Bagheera kiplingi, and as far as I can tell it has no common name. It lives primarily off nutrient-rich nubs found at the base of leaves of Mimosaceae trees called Beltian bodies, which normally incentivize symbiotic ants to colonize the tree and guard it from other predators. The ants protect the tree against, among other things, B. kiplingi. The spider also drinks nectar, and will steal ant larva and cannibalize other B. kiplingi if given the opportunity (hence, it not quite being herbivorous).

January 08, 2016 at 07:29AM

A bit late, but without further ado: Today I Learned: 1) Apparently just about everything I know about evolutionary psychology (taht is, the evolution of behaviors) comes from a scientist I'd heard of, but never really given anything other than a last name in my mental indexing scheme, by the name of Robert Trivers. He came up with the textbook evolutionary theories for reciprocal altruism, sex differences, cheating, cooperation, sexual jealousy, self-deception, parent-child conflict, romantic bonds, and others. Robert Trivers also seems to be one of the more, uh, real people on the planet. In addition to winning one of the most prestigeous worldwide prizes in evolutionary theory, he, I quote, "was a member of the Black Panthers and collaborated with the group’s founder. He was arrested for assault after breaking up a domestic dispute. He faced machete-wielding burglers who broke into his home and stabbed one in the neck. He was imprisoned for 10 days over a contested hotel charge. And two men once held guns to his head in a Caribbean club that doubled as a brothel." That sounds like a full life to me! Thanks to Michael LeFew!!!! for linking me to this fantastic little article on Robert Trivers: http://ift.tt/1ZP0dUC 2) There are order-of-magnitude around 1,000 publications a year on RNA-seq. 3) There are somewhere around 300-350 "moderate" to "high" publication rate research universities in the US, according to Certain Metrics (TM).

January 06, 2016 at 10:15PM

Today I Learned: 1) Near the Salton Sea (which I talked about in my last TIL) is a place called Slab city. Slab city is an abandoned marine bunker facility, now inhabited part-time by (mostly very poor) snowbirds and full-time by a core community of very dedicated folks. The city ("city", really) has no electricity, sanitation, running water, or, as far as I can tell, government of any sort, but it makes up for the lack of these in art. Thanks to Lady Jade on this one! 2) There's an alternative to non-homologous end joining* called microhomology-mediated end joining, or MMEJ. MMEJ can occur when double-stranded DNA breaks with short, matching sequences (microhomologies) nearby on each side. The cell can find these microhomologies, pair them together, and fill in any gaps. This process deletes any sequence between the break and the microhomologies, making it a destructive repair process, and MMEJ is a common cause of cancer-inducing mutations. Why would we have a repair pathway so prone to causing cancer? I really don't know. I mean, destructively repairing a double-stranded break is usually better than the cell just dying, which is what happens if you leave a break un-stitched (which itself is probably better than trying to replicate with a double-stranded break... that could cause all sorts of havoc). Maybe it's a last-ditch effort to patch things up in damaged cells before replication? Whatever the reason for its existence, MMEJ can be used for genomic engineering. Check out http://ift.tt/1OPGBYr for details, if you have a Nature subscription. Credit to the Addgene Facebook page on this one. * The most common process cells use to stitch broken DNA back together. 3) You can make chefs knives out of bamboo. Anyone have experience with these? Andrew Andy Halleran? Thanks to Mengsha Gong Bonus) Sharks. In volcanos. https://www.youtube.com/watch?v=0e3t18rrjOA (Thanks again Mengsha Gong)

Huxley's Other Book, The Salton Sea, and Endorheic

Today I Learned:

1) Aldous Huxley, author of Brave New World, wrote another book called The Island which was about a future civilization that used the same techniques and technologies as those used in Brave New World, but to utopic effect rather than dystopic effect. I have somewhat more respect for Aldous Huxley, and I suspect the two novels should really be taught together (unless The Island just isn't any good, which is possible).

I got this from somebody on Facebook this morning, but I have to confess I'm not sure who it was, looking back. Whoever it was, thank you!

2) The Salton Sea is an interesting body of water in the southern California desert. The Salton Sea is an inland (and endorheic -- see below) body of water that actually goes through 4-5 century cycles of freshwater lake -> saltwater lake -> desert. Right now it should be in the desert phase, except that a miscalculation in some hydrological engineering in the early 20th century flooded it early. Since then, it has been steadily salinifying and moving around. If you're interested in dynamic geology, check out the Salton Sea.

Thanks to Tara Sullivan for posting some intriguing pictures of the Salton Sea!

3) "Endorheic", as in an endorheic lake or water basin, is a body of water that doesn't empty into an ocean or sea. This isn't the first time I've run across the root "rheos" in my TILs. Turns out it's Greek for "flow".

Superhiders, MMO Purchase Rates, and Jarvis

Today I Learned:

1) You know how you can tell Facebook to hide a post and not show other posts like it? It's an important tool for Facebook to learn what users don't want to see, and what you, in particular, don't like to see. Today I learned that there's an important outlier in hiding that skews Facebook's data significantly -- it turns out that a very small fraction of the Facebook userbase uses the "hide" feature on *everything*, including things they read and like. These "superhiders" use the hide feature to mark which posts they've read and get them out of the way, like an archiving system. They also account for a majority of Facebook's hiding data, which rather skews the company's feedback.

More info (from Facebook) at http://ift.tt/1OIyVak.

2) MMO developers typically plan on 1-5% purchase rates for optional purchasable items. I'm sure it varies by game and item, but it's a nice order-of-magnitude number to know for calculating MMO game incomes.

3) Mark Zuckerberg has publically announced his New Year's resolution for 2016 -- he's going to write an AI to help manage his home, a la Jarvis. Thanks Mark Zuckerberg!

Element Names, Beaver Bites, and Delicious Drinks

Today I Learned:

1) The IUPAC has specific criteria for what you can name a new element -- it can be a "mythological concept, a mineral, a place or country, a property, or a scientist". Elements are named by the teams that discover them, although I don't entirely know what it means to discover an element (I gather it's really more synthesis of the element -- for short-lived elements it seems to be unequivocal detection and identification of the breakdown products). More info about naming elements at http://ift.tt/1kDTDjV.

2) What beaver-bitten trees look like, in person. I think. I don't know what else those could have been.

3) Hershey's chocolate syrup and soy milk make a delicious, delicious drink. (Also, from last night -- Frangelico and soy milk).

Brittle Quartz, Space Crystals, and The Most Popular Plasmid

Today I Learned:
1) Even rock that looks like quartz is sometimes quite brittle -- in the case of the one I tested, much more brittle than a fist-diameter branch of wood.

2) Crystals can be grown larger in space than on Earth. It turns out that gravity-driven sedimentation is an important limiting factor on crystal growth planetside, which obviously isn't a problem in space.

Why would we want to make bigger crysals? Well, aside from giant crystals being cool (have you *seen* the stuff in the Smithsonian's crystal collection?), one of the big problems in biology/biochemistry is the problem of crystallizing proteins, which is required for X-ray crystallography, which is the only way to definitively see the shape of a protein. Apparently some proteins that couldn't be crystallographed on Earth have been crystallized on the ISS, allowing their analysis.

3) The most-sold plasmid on Addgene (the primary source of plasmids in academia) was lentiCRISPRv2, a plasmid for building lentiviral vectors for cas9 (and a guideRNA). Congratulations to Feng Zhang's laboratory! (For scale, lentiCRISPRv2 was ordered around 1,000 times this year, according to addgene. All things considered, academia isn't a huge market....)