April 12, 2016 at 11:52PM

Today I Learned: 1) The list of games Buddha would not play: http://ift.tt/1TQkicg This list is fascinating on many levels. On one level, it's an interesting historical glimpse into what kinds of games may have been played in ancient India. It's interesting how familiar most of these games are -- there seem to be some more-or-less universals in the human condition regarding games. On another level, it's fascinating that Buddha would choose to write about not playing certain games, and there are a lot of possible impliciations for the list. Is this Buddha saying that he refuses to play games, a la Green Eggs and Ham? "I will not play them in a house, I will not play them with a mouse. I will not play them here or there -- I will not play them anywhere!" Or perhaps these are games the Buddha would not play because they specifically care about things the Buddha doesn't care about. I can think of possible reasons to not play most or all of the games on the list -- games on boards with rows and columns tend to be all about strategizing under man-made rules; ball-games are all about physicality; playing with toy versions of real things is kind of ignoring the reality of the real thing; imitating deformities is not a good way to cultivate compassion. Finally, there's #6 -- "Dipping the hand with the fingers stretched out in lac, or red dye, or flour-water, and striking the wet hand on the ground or on a wall, calling out 'What shall it be?' and showing the form required--elephants, horses, &c." Really? I mean, who wouldn't! 2) ...how to digital qPCR! It's remarkably easy, actually, given how much better it seems to be than old-fashioned qPCR. Chemically speaking, it basically *is* qPCR, except the master mix is different, and it's much more robust. The biggest difference is that there are a couple of extra mechanical steps. For the uninitiated, digital qPCR is a technique for quantifying amounts of DNA in a sample with very high accuracy. You basically make a qPCR mix (Taq polymerase, buffers, probes, and some sort of DNA-binding dye), use a microfluidic device to split the mix into about 20,000 microbubbles in oil (much, much, much easier than it sounds, using a disposable microfluidic device) and thermal cycle the whole thing to completion. The goal is to have the sample so dilute that you get approximately one copy of DNA per bubble. That way, some bubbles will randomly get one copy, some will get zero, and some will get a bit more. When you run the PCR reaction, each bubble will act as its own little PCR sample, and will amplify up a ton of DNA if there was one or more copies of the target in that bubble. Then you use a special bubble-imaging machine to count the bubbles with and without amplified product -- anything with amplified product will be brightly fluorescent (from the DNA-binding dye), and anything without amplified product will be essentially blank. From the counts of bubbles-with-no-product and bubbles-with-product, you can back-calculate the starting concentration of DNA. The nice thing about digital qPCR is that it uses binary endpoint-PCR results -- if there's product, you get a hit, and if there isn't, you don't, hence "digital". This makes it super insensitive to stuff like primer binding, salt conditions, inhibitors, and other kinds of things that could mess up traditional qPCR results. Another nice thing is that the readout you get at the end is an absolute quantity, not a relative quantitation, so you don't have to run every sample with a ladder to quantify how much stuff you have. Thanks to Erik Jue for getting me in on some dqPCR training! 3) When using pip (or conda) on a UNIX-like machine, make sure you have superuser priviliges (i.e., sudo) any time you use it to upgrade itself! Otherwise, it will uninstall the old pip(/conda), download the new one, and fail when it tries to install without superuser priviliges, leaving you without a working pip(/conda). Fortunately, pip is pretty easy to install with a bootstrap script (http://ift.tt/1mn7OFn), so if you *do* screw yourself over this way, it's relatively easy to recover from.