2015-07-14.log

--- Log opened Tue Jul 14 00:00:10 2015
fenni dont see any fume hood at ccl but there are a couple small rooms if that's relevant and there is access to the ceiling00:07
-!- Porb [~Porbus@CPE-124-176-219-129.lns5.win.bigpond.net.au] has joined ##hplusroadmap00:10
-!- Acty [uid89656@gateway/web/irccloud.com/x-bkperneclkheqwyj] has joined ##hplusroadmap00:11
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Read error: Connection reset by peer]00:16
-!- wrldpc1 [~ben@206.104.133.27.ap.yournet.ne.jp] has quit [Ping timeout: 246 seconds]00:19
fennlooks like they are planning on putting in a fume hood00:21
-!- Porb [~Porbus@CPE-124-176-219-129.lns5.win.bigpond.net.au] has quit [Quit: Leaving]00:45
-!- proofoflogic [uid65184@gateway/web/irccloud.com/x-tqtjxumfgrwrjxjd] has joined ##hplusroadmap01:07
juulfenn: you should come by for our tuesday social hangouts. beer and pizza 7 pm01:39
-!- zadock [~outsider@cthulhu.tuiasi.ro] has joined ##hplusroadmap01:42
fennder.. i'll see if i can make it01:51
-!- zadock [~outsider@cthulhu.tuiasi.ro] has quit [Quit: Leaving]01:58
-!- JimmySRussel [~Razk@unaffiliated/knobuddy] has quit [Remote host closed the connection]02:22
-!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap02:58
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 264 seconds]02:59
-!- Viper168_ is now known as Viper16803:20
-!- proofoflogic [uid65184@gateway/web/irccloud.com/x-tqtjxumfgrwrjxjd] has quit [Quit: Connection closed for inactivity]03:27
kanzuregood03:45
kanzure23:11 <xentrac> so 1.2MHz/5 gives you a limit of 240 kilobases synthesized per second, at which point you want your conveyor belt to be about 28 million pixels long; at 1200dpi, which is probably optimistic, that's 609 meters, or 6.8 meters per nozzle03:56
kanzure23:12 <xentrac> (the 28.8 million pixels is 2 minutes)03:56
kanzurehmm apparently people have cnc machined plutonium. i want to do this.03:59
kanzurestrange, iran just agreed to be banned from making nuclear explosion simulations04:04
kanzureand also banned from making streak cameras... why would they agree to that.04:05
fennbecause nudity is morally reprehensible04:21
-!- drewbot_ [~cinch@ec2-54-226-221-122.compute-1.amazonaws.com] has quit [Remote host closed the connection]04:21
-!- drewbot [~cinch@ec2-54-158-55-149.compute-1.amazonaws.com] has joined ##hplusroadmap04:21
-!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has quit [Quit: Beatzebub]04:45
fenni'm a bit put off by azco's totally lame and incomplete MSDS for phosphoramidites04:54
fennPotential Health Effects (Acute and Chronic): No data available.   -_-04:56
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap04:57
-!- nottimschmidt [~timschmid@2601:405:4101:52db::822] has joined ##hplusroadmap05:02
-!- _hanhart [~hanhart@static.101.25.4.46.clients.your-server.de] has joined ##hplusroadmap05:21
-!- eudoxia [~eudoxia@r167-57-2-102.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap05:26
fennkanzure i will send that email to counterculture unless you have any additions/subtractions05:27
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap05:47
-!- nottimschmidt [~timschmid@2601:405:4101:52db::822] has quit [Ping timeout: 248 seconds]05:49
-!- narwh4l [~michael@unaffiliated/thesnark] has joined ##hplusroadmap06:06
ebowdenpaperbot http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=78481606:09
juri_nmz787: is there an open toolchain for programming the ICE40?06:11
-!- nottimschmidt [~timschmid@35.10.220.154] has joined ##hplusroadmap06:43
-!- Madplatypus [uid19957@gateway/web/irccloud.com/x-dvyjjsdompilywnq] has quit [Quit: Connection closed for inactivity]06:46
CaptHindsightjust out of curiosity why would some want to spend days or weeks to develop new FPGA's, drivers or even an application to replace something that already works really well and costs <$200?06:48
CaptHindsightand only panning on building a few06:49
juri_someone who has standardized on a different solution?06:51
nottimschmidt< $200 might as well be infinitely expensive for most of the world's population.06:52
juri_I try to cut costs on every part of the things i build, for the reason nottimschmidt just supplied.06:53
juri_one of my 3d printers uses a microcontroller salvaged from a battery backup i found on the side of the road. because.06:53
nottimschmidtIn the world of collaborative development, fewer barriers to entry means more contributors :D06:54
CaptHindsightnottimschmidt: the <$200 in this example is part of a $10K machine/system06:54
narwh4lIt's important to distinguish open design from a hacked design06:54
narwh4lopen designs are great for developing countries. Not sure how accessible the microcontroller from a battery backup is to the world06:54
CaptHindsightso if you can't afford the $10K it's not an issue anyway06:55
nottimschmidtCaptHindsight: I don't think this philosophy or way of looking at things should prevent you from doing the work in the way that works best for you.  But you should be aware of the motivation of others in reducing cost and complexity everywhere we can.06:55
nottimschmidtToday's $10k machine is tomorrow's $1k machine06:55
nottimschmidt3D printers are already available in the $300 price range06:55
nottimschmidtThat happened one cost reduction at a time :D06:55
fennit's also easy to say "hey i can buy a printer for $50 so why is this $10k"06:56
nottimschmidtyep06:56
juri_it's also important to note that today's 3d printer has much better software than yesterday's $1K printer.06:58
nottimschmidtand if that software is well architected, it serves as a library of modular components that makes building the next machine that much easier :D07:06
fennthanks for your work on transfering things to git btw07:07
nottimschmidtThe work I've done with SLS, SLA, and laser cutting was all easier thanks to the 3D printer software available.07:07
nottimschmidtfenn: me or someone else?07:13
nottimschmidtIt's been a while since I moved anything to git :D07:13
fenniirc you moved all the reprap SVN stuff and converted it to .scad files07:13
nottimschmidtYeah.  You have a long memory.  :)07:14
fennit was a big mess because they just threw everything into svn regardless whether it was relevant or not07:14
fenni think if that hadn't happened nobody would have been able to make improvements to any reprap software07:15
nottimschmidtThanks for that.  Lots of other folks contributed as well.  These days I'm thinking a lot about ways to make the firmware and CAD more flexible.  Supporting lots of different printable and off-the-shelf tool holders/changers in CAD and the firmware, porting things to javascript, and trying to psych myself into writing a toolpather based on this: http://dspace.mit.edu/bitstream/handle/1721.1/29225/50140264-MIT.pdf?sequence=207:19
fenn.title07:20
yoleauxfenn: Sorry, that doesn't appear to be an HTML page.07:20
nottimschmidtpdf07:20
nottimschmidtIt's a masters thesis paper that outlines all the algorithms necessary for automatic 5 axis toolpath generation07:21
fenn"automatic 5-axis NC toolpath generation by mahadevan balasubramaniam"07:21
nottimschmidtThere's currently no Free software capable of that07:21
CaptHindsightheeks and pycam only do 3-axis07:23
CaptHindsightnottimschmidt: are you going to write a 5-axis CAM app as well?07:23
nottimschmidtCaptHindsight: we'll see.  I've been hacking a bit on OpenJSCAD.  I like that I can run it on my phone.  Most of the world computes with phones.  I'd like to teach it to slice and print.07:24
-!- knobuddy [~Neff@unaffiliated/knobuddy] has joined ##hplusroadmap07:26
nottimschmidtI'd like to teach it about multiple tools.07:26
nottimschmidtBut I am currently teaching it about CAD libraries.07:27
nottimschmidtSo it's easier to make designs from an assembly of pre-built parts, instead of having to build everything from scratch.07:27
fennis pycam usable in general now?07:28
kanzurefenn: email looks good to me07:29
nottimschmidtWe've used it a bit, for the laser cutter.  Seems to mostly work, if not fully.07:29
nottimschmidtSeveral successful cuts.07:29
nottimschmidtThis is the tech I'm working toward: http://us.dmgmori.com/products/lasertec/lasertec-additivemanufacturing/lasertec-65-3d#Video07:32
nottimschmidtI can't afford to buy one, but I want one :D07:33
fenni remember this, very cool07:35
fenna tig welder could probably do just as well for the additive part07:36
kanzurefenn you might be interested in where nottim isn't working these days07:36
fennerror negative stack size exceeded07:36
nottimschmidtfenn: true.  I like the part lifecycle of the powder printing - recycle failed prints in a ball mill, re-print.  But a MIG/TIG would work just as well for v1.  Good idea.07:37
kanzurehttp://beacon-center.org/07:37
nottimschmidtyep, that place.07:37
fenndo you make beacons07:38
nottimschmidtI have been absorbed by the warm and comforting robot bosom of the academic borg.07:38
kanzureevolutionary search algorithm stuff07:38
kanzurenottimschmidt: have i shown you http://verbnurbs.com/07:39
nottimschmidtBiologists, computer scientists, physicists, ecologists, and so on.  Studying computational biology, intelligence, genetic programming, evolutionary algorithms, etc etc etc.07:39
nottimschmidtkanzure: yep!  I'm still looking for exactly the right use.07:40
fennevolutionary algorithms is a big hammer07:40
nottimschmidtyep07:40
nottimschmidtThere's a group here doing "On Intelligence" style research as well.  Using markov brains, c. elegans models, etc.07:42
nottimschmidteven bigger hammer07:42
fenni don't know what a markov brain is07:42
nottimschmidtyeah, no one does.07:43
kanzureperfect07:43
nottimschmidtI don't think anything's been published yet07:43
fennquick, race to the patent office!07:43
fenn"yes hello i'm trying to patent the concept of a markov brain. what is it, you ask? well we don't really know but it sounds cool and we want to prevent anyone else from using it for 20 years"07:44
fenni think "long short term memory" and "recurrent neural networks" are the new hot topics07:45
fennbasically the same thing07:45
nottimschmidtThey're really simple.  Imagine an area of memory a few bytes long.  Say 8 of the bits in that memory are dedicated to inputs - one bit represents an eye, another bit represents another eye, say one represents a food sensor, etc.  And a few more bits represent motor outputs.  move forward / back for a left wheel and a right wheel.07:45
nottimschmidtConnect those bits with an evolved network of logic gates.07:45
nottimschmidtHierarchical Temporal Memories are the shit07:46
nottimschmidtMarkov brains can be functionally identical, just another way of expressing the same algorithms.07:46
fennboolean logic? is there a delay in the transmission from one gate to the next?07:46
fennin order to make use of rate coding you have to have some kind of accumulator07:47
fenneh well anyway, i hope it works out07:48
nottimschmidtNo delay, but determistic, incremental time.  Depends on the implementation, but gates can function as memory, as can bits in the systems memory that are connectable to gates, but don't function as input or output07:48
fenna long time ago someone did experiments with evolved fpga code, worked beautifully and hardly used any gates at all, but it wouldn't run in simulation or on any other chips07:48
nottimschmidtThere are feedback gates in some implementations as well07:48
nottimschmidtRight07:48
nottimschmidtThese guys to really interesting things in very few gates as well, but the behavior is deterministic.07:49
nottimschmidtOne researcher I know is working on calculating Phi for these things07:50
nottimschmidthttps://en.wikipedia.org/wiki/Integrated_information_theory07:50
nottimschmidtPhi's a measure of the amount of information integrated by a system07:51
kanzure.wik07:51
yoleauxSearch for an article on Wikipedia07:51
kanzure.wik integrated information theory07:51
yoleaux"Integrated information theory (IIT) is a framework intended to understand and explain the nature of consciousness. It was developed by psychiatrist and neuroscientist Giulio Tononi of the University of Wisconsin–Madison." — https://en.wikipedia.org/wiki/Integrated_information_theory07:51
kanzurebleh07:51
kanzurehere's an explanation of consciousness: it doesn't exist. there, problem solved.07:51
nottimschmidt:D07:51
nottimschmidtPrediction is what matters.07:51
nottimschmidtAnd is how the local physicist defines information07:51
kanzureno really, what predictive benefit does consciousness confer on your models that is impossible without consciousness?07:53
nottimschmidtI'm agreeing with you.  :D  I think we're memorization and prediction machines, and that's the bulk of what constitutes intelligence.07:54
kanzurei also don't believe in the intelligence word either07:55
fennthey ain't no sich word07:56
nottimschmidtI think that's fair.  My understanding is an ecological one, at several different scales.07:57
nottimschmidtEmergent properties of complex interacting systems, all the way down :D07:57
nottimschmidtAnd all words and names for such things somewhat inadequate and colored extensively by perception07:58
fennperfect denotation is raw data; all vocabulary carries compression artifacts07:59
nottimschmidtyup07:59
kanzurei'm not complaining about compression artifacts07:59
kanzurei'm complaining about wrong or broken concepts07:59
fenni don't think consciousness is "wrong" as a concept, it's just "not even wrong"08:00
nottimschmidtkanzure: they are everywhere, and fuck up understanding like a damn fucks up a river.08:00
nottimschmidtbrb08:01
kanzurei'm not claiming there are no compression artifacts flying around08:01
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Ping timeout: 255 seconds]08:12
kanzurefenn: email sent?08:12
-!- Porb [~Porbus@CPE-124-176-219-129.lns5.win.bigpond.net.au] has joined ##hplusroadmap08:17
fennsented08:18
fennoh i should have cc'd you so the replies went to you too08:18
fennoh well08:18
kanzurewell depending on how much you care about that, you can send a follow up where you cc me and say "looping bryan into this"08:19
-!- Porb [~Porbus@CPE-124-176-219-129.lns5.win.bigpond.net.au] has quit [Ping timeout: 256 seconds]08:51
-!- nmz787_i [~ntmccork@134.134.139.77] has joined ##hplusroadmap08:51
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-knzxqmqtwcqwsjcp] has quit [Quit: Connection closed for inactivity]08:58
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 246 seconds]09:01
nmz787_ijuri_: http://hackaday.com/2015/05/29/an-open-source-toolchain-for-ice40-fpgas/09:04
juri_nmz787_i: Nice.09:05
nmz787_iCaptHindsight: I'm all for splurging and dropping $$$ when I want to be lazy and not spend a ton of time engineering something... but I also like to have a path forward for thinking of how to decrease costs. I think the others gave lots of good points too. I used to not have lots of $, but still knew that for a lot of things, the parts/materials cost was significantly less than whatever sales pricetag something had slapped on it. $200 doesn't09:08
nmz787_isound too bad, but I'm pretty sure it could be 1/5th of that or less with some customized boards... or even just picking and choosing from existing boards that are up on ebay, etc... (i.e. ice40 and an easydriver)09:08
nmz787_iCaptHindsight: if for no other reason, I like to have relative comparisons just for judgement and keeping my bearings in a space09:09
juri_this makes some good sense. it's worth noting the ICE toolchais speaks verilog, and the mesa toolchain is VHDL.09:10
nmz787_iah, how about the relative size of the fabric in that compared to the mesa?09:17
CaptHindsightjuri_: does that even matter?09:17
CaptHindsightseems some people would spend weeks hand packing gates to save $2 since their time must be free09:18
juri_CaptHindsight: you're right in a fashion.09:19
CaptHindsightif you are going to mass produce it sure09:19
juri_saving $2 is a big dial if you're expecting people to build hundreds of them.09:19
kanzureheath: you should go hang out with fenn09:20
CaptHindsight1,000,000 x $2 is a lot of $09:20
kanzureheath: counterculture labs has 7pm social gathering09:20
juri_right.09:20
nmz787_iCaptHindsight: those people seem crazy... the largest countries still make $1 a day... so their time is worth more than $2 per weeks09:20
CaptHindsightbut cnc glue guns and an inkjet printer are not going to revolutionize their world09:22
juri_some hackers are very poor, despite being in rich areas.09:22
-!- nmz787_i [~ntmccork@134.134.139.77] has quit [Quit: Leaving.]09:24
CaptHindsightI often see lots of reinventing for the sake of reinventing09:25
CaptHindsightespecially with user interfaces09:26
CaptHindsightwhy do they all seem to get worse?09:26
juri_we all have radically different UI philosophies. for instance, i prefer writing my code in emacs and building it using Makefiles.. including my 3d models.09:34
kanzurethis just sounds like you guys are angry about spending money on stuff09:35
kanzureCaptHindsight: we're investigating counterculture labs as a possible home for the machine09:38
CaptHindsightare they in Oakland or the area?09:39
kanzureoakland09:41
eudoxiaanyone here remembers that post in mike darwin's blog where he talked about paying a master glass blower to make a bubble trap09:42
eudoxiai recall it had a tiny insight about the value of apprenticeship09:42
fennit's all the bay area09:42
fennoakland is just known as "oakland" because it sucks09:42
fennand maybe something about a football team...09:43
kanzure"Also, he really cared about the ant hill more than the ants. He would do horrible things to the ants that would encourage them to build a bigger ant hill. (Ants are users, employees were treated very well). As a fictional example, lets say the site was skewing too heavily male by 20%, just add a line in the stored proc that gave a male user a 20% chance of their registration not being written to disk. Voila, site has the proper mix, ...09:45
kanzure... onto the next problem. He was the most brutally effective person I've ever worked with."09:45
kanzureeudoxia: talked with mike darwin for about 3 hours on the phone the other day. i have prepared for future calls and will record things in more detail.09:45
eudoxiakanzure: cool, what did you talk about09:46
kanzurebottom of http://gnusha.org/logs/2015-07-11.log09:47
-!- nmz787_i [ntmccork@nat/intel/x-iqdutpawaebcfcug] has joined ##hplusroadmap09:50
kanzurefenn: ah, heath reports that he will be attending counter culture labs this evening, so perhaps you will run into him if you go09:52
fenneleitl suggested literally recording the sound of the conversation09:52
kanzurewhich conversation?09:53
-!- nmz787_i1 [~ntmccork@134.134.139.78] has joined ##hplusroadmap09:53
fennugh so much obligation stress.. now i definitely will not be able to get to sleep09:53
fenni am not going, ok, good.09:53
kanzureobligation stress eliminated by planning to not go?09:53
fennsomething like that09:53
fenni wish it were so simple09:54
kanzurewell, you sent the email, if they aren't raging assholes then they should consider the proposal as a group09:54
-!- nmz787_i [ntmccork@nat/intel/x-iqdutpawaebcfcug] has quit [Ping timeout: 256 seconds]09:54
nmz787_i1i just left?09:54
fennit's basically impossible for me to fall asleep when i know i have to wake up at a certain time09:54
nmz787_i1oh, no, i am the clone09:54
kanzurefenn: really the root of the problem here is not obligation stress or sleep schedule mismatch but that they are having an in-person meeting (it shouldn't be)09:55
fennit's not a meeting, it's just beer and pizza09:56
kanzureoh hm09:56
kanzureyeah i guess that's difficult to orchestrate without bodies in the same room09:56
fennyou could use a clever system of catapults09:57
kanzurei think you run out of available pcr barcodes09:58
fennanyway you should record your phone conversations with mike darwin for posterity09:58
kanzureand if you increase the size of the barcode then the middles of the barcodes start matching the ends of other barcodes because of sequence similarity09:58
kanzureyes, i understand09:58
fennthe length of the oligo is limited anyway09:58
kanzurenumber of spots is not limited09:59
kanzurexentrac wants a conveyor belt added heh09:59
fennthe barcode is part of the oligo sequence09:59
fenni dont know what you guys are intending to do with all this dna09:59
kanzurei'm not sure whether to think that single pool gene assembly is going to work10:00
kanzurewell, i'd like plasmid and gene assembly10:00
kanzureand genome assembly would be nice10:00
juri_fenn: replicants.10:00
fennwell it's not really single pool if you're selecting different subsets via pcr and then doing assembly in a separate reaction vessel10:00
kanzureok, so you pcr-amplify in the same pot, then you extract the ones that you amplified (because most of the extractant is going to be your amplified material)?10:01
fennok so how big is a plasmid, 100kB at most right?10:01
kanzuresure 100 kb sounds good to me10:01
fennwell there's orders of magnitude left over if you're doing millions of oligos10:01
CaptHindsighthow do you think that they are doing assembly? http://www.adnas.com/products/signaturedna10:02
kanzurethis doesn't say anything about gene assembly10:03
fenn"SigNature DNA markers are based on full, double-stranded plant DNA."10:03
nmz787_i1100kb is huge for a plasmid10:03
heathfenn: are you planning attending the cc social?10:03
nmz787_i1more like 40kb max10:03
fennlol "This botanically engineered solution is shielded by a portfolio of 24 patents, 58 patent applications, and other intellectual property protection."10:04
-!- Madplatypus [uid19957@gateway/web/irccloud.com/x-hfvamhwbbeebqzld] has joined ##hplusroadmap10:04
nmz787_i1heh, proxyham http://hackaday.com/2015/07/14/how-to-build-a-proxyham-despite-a-cancelled-defcon-talk/10:04
CaptHindsightheh yeah10:04
fenn"we applied for so many patents it hurts"10:04
kanzurenmz787_i1: ah didn't know there were 40 kb plasmids. cool.10:04
nmz787_i1fenn seems to want some proxypizza10:04
kanzureit wasn't a ham10:05
kanzureit was lies10:05
kanzureheath: i have sent you an email that fenn sent earlier to counter culture labs10:05
nmz787_i1kanzure: hmm, from nets "Yes, the bigger the plasmid the lower the transformation efficiency. However, plasmids up to 200 kb can be transferred by electroporation (hand-on experience). For even bigger plasmids conjugation would be my choice of transfer method."10:05
kanzurecool, glad to hear 200 kb plasmids can get in there10:06
nmz787_i1kanzure: I don't see that email on the mailing list... was it not sent that route?10:06
kanzurewhich mailing list?10:06
nmz787_i1and also "Plasmid stability/ segregation and ease of work. It simply gets more difficult to handle when they are large (> 50 kb)."10:06
nmz787_i1the counterculture mailing list (google)10:06
kanzureno it was not sent to that email address10:06
fenni am lame and haven't looked into mailing lists or anything yet10:08
fenni'd rather talk to a specific person than "hey guyz how bout this thing ... *crickets*"10:08
fenn"All you need to do is cancel the talk and allow tech journos to speculate about National Security Letters and objections to the publication of ProxyHam from the highest echelons of government."10:10
fenni dont get it anyway... 900 MHz? just use a wifi bridge10:10
kanzuresee https://www.reddit.com/r/worldnews/comments/3d7wil/a_project_to_build_a_200_diy_wifi_router_to_help/ct2tvn610:10
kanzurei think that making a vitamin a producing probiotic would be a good project to do with this dnajet10:13
nmz787_i1kanzure might say you could accomplish internet anonymity with a clever catapult setup, throwing zip disks around with encrypted data10:13
kanzureunfortunately that suffers from man in the middle attacks10:13
fennalso disk read head oxidation10:14
fennas anyone who has tried to read a zip disk knows10:14
kanzure"Sure, I'll plan on being there and speaking with Kathy if I don't see Fenn."10:15
fennthat commenter goes a bit too far calling 900MHz "crap" - it's only 2.6x slower than 2.4GHz and usually you're nowhere near saturating the wifi link10:16
fennbut it does go through trees and fog better10:16
kanzurepick-and-place seems to be the only gene assembly technique available at the moment. you have to limit the number of beads, there's no "assemble as many things as you want simultaneously" technique that anyone has figured out.10:24
kanzureactually, a method that would be nice would be something like: imagine a giant grid where you print all your spots. there are separate groups of spots that will (somehow) be mixed together after initial synthesis. these groups will undergo some chemical reaction for local assembly. then these groups can be mixed together (without pick-and-place), and so on, up to at least 3 or 4 levels.10:26
fennthe trick is to not do it simultaneously10:26
kanzurethis could perhaps be achieved by something like, "use a laser to cut some open-top channels between the groups, so that the liquids can mix later, and then you just pcr in the combined droplets"10:26
fennthe pick and place is for error rejection mostly10:26
kanzureit helps for that, but no i think it's so that you can pick which 10 beads you want to run gibson assembly on10:27
kanzureand then you do a 10-bead gibson assembly run in another test tube (or whatever), then you combine the two results and run gibson assembly again10:27
fennyes the 10 beads that have no errors :P10:27
kanzurequality control is a separate question i think10:28
kanzurethe 10 beads would have separate sequences (intentionally)10:28
kanzurewith the pcr barcodes10:28
kanzurelaser cutting the surface to mix the different groups might work10:29
kanzureplus the same laser could perform laser-assisted-heating pcr10:29
kanzureyou could also use stencils to confine different groups, then remove a stencil to allow mixing, and then remove the next stencil to allow the next bracket to fight10:30
kanzureerr to allow the next bracket to assemble10:30
nmz787_i1if you were enriching via the PCR barcodes for a million spots, wouldn't you need a million primers in some storage area10:31
nmz787_i1?10:31
nmz787_i1and I think you could just jet mastermix between the spots and join them together with surface tension10:31
kanzurehow big can the spots get before surface tension is not enough10:31
kanzureah you can add stuff to make surface tension to continue to work i think10:32
nmz787_i1if you had 4 spots, jet an X of mastermix+enzymes10:32
kanzurehmm10:32
kanzuresurfactants10:32
kanzurejoining the spots together is an interesting idea10:32
kanzureyou would have to lay out your dna bitmap so that you can combine the spots you want in the correct order for assembly, of course10:32
nmz787_i1well surface tension works for spots millimeters wide (at least thinking of my car window when it rains) so I'm guessing it's only stronger for sub-millimeter10:33
nmz787_i1for a million spots, how many mers do you need?10:33
nmz787_i1is that log4(1000,000)?10:33
nmz787_i1yea10:34
nmz787_i1so 10-mer10:34
nmz787_i111 gets you 4 million10:34
nmz787_i1unless the spots are too low conc to amplify later, after ligation and transfer losses.... then I'd say skip the barcodes... rely on hybridization of gibson assembly, etc... then after a few pooled assemblies, amplify with some generic starting barcode and filter by length (by that time there should be an order of magnitude difference between synthesized and amplified)10:37
nmz787_i1(length)10:37
fennnmz787_i1: you dont need a million different primers, only however many subsets you have. if it's 10 oligos per assembly then you need 100,000 primers10:40
kanzuremoving droplets wouldn't work if you need the wash step to not wash away everything, plus if you're using pores or wells then you need to figure out some other way of mixing between liquid-bridged pores i guess.10:42
kanzuremoving/merging10:42
nmz787_i1fenn: kanzure wants 1.2 millions spots, if they're all unique... won't you want each to have it's own barcode? anyway, if they're destined for assembly, hybridization is kind of the same thing... so do what I said and I think it will be easier10:46
fennit depends what you're using the barcode for10:47
nmz787_i1wasn't it supposed to be like a checksum, only good codes hybridize with the primer and amplify?10:47
fennone system used the barcode to pick out randomly labeled clones of individual oligos. in that case you'd need a million barcodes (or however many oligos you wanted)10:47
fennanother system used a barcode to pick out a subset of oligos for hybridization10:48
fenner, assembly, not hybridization10:48
fennit's probably better to use smaller oligos anyway because parallel writes go faster than serial writes10:49
fennit takes 10 times longer to make an oligo 10 times as long, but less than that to make 10 times as many oligos10:50
fennlike 1.01 times as long maybe10:51
nmz787_i1maybe you could use 2 barcodes, one at the beginning and one at the end... but then you need to clip them off after amplification and it doesn't ensure there aren't internal deletions10:51
nmz787_i1which hybridization does help with10:51
kanzurewhy would you need to clip them off after amplification10:51
fennto ensure they are random sequences10:51
nmz787_i1how will your protein function with some repetitive tag every 30 bases or whatever10:52
nmz787_i1you can't have that10:52
kanzurei think proteins could tolerate that, actually10:52
kanzureespecially if it's not an amino acid10:52
fenner.. every sequence is an amino acid10:52
kanzurehmph10:52
nmz787_i1i don't know ascii emoticons well enough to describe how I just felt about that comment10:52
kanzureright, right10:52
kanzurei'll go back to writing software10:52
kanzureduring pcr and assembly perhaps you don't need wash steps10:56
fennnow that i think about it, it might be difficult to do pcr on a 12"x12" plate10:59
fennwater evaporating and stuff10:59
kanzureneutral atmosphere, humidity could be modified if necessary11:00
nmz787_i1might need to be a separate machine too, since you don't want water near the synthesis stuff11:00
fennthis is after the plate is removed11:00
kanzure.title http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3209801/11:00
yoleauxPetri dish PCR: laser-heated reactions in nanoliter droplet arrays11:01
nmz787_i1if you contaminated the inert zone, you'd need to pull vacuum for a while to dry things out11:01
fennmaybe it makes sense to be able to easily break the plate into chips, so you can just put a chip with 10000 oligos on it into a pcr tube instead of messing with a whole plate11:02
kanzurei'm not convinced anyone has a method of assembling a gene from 10k oligos11:03
fennwhat was the density again? in oligos per mm^211:03
kanzurethat was undecided actually. i think at one point we had 100 microns between drops, possibly more.11:03
fennyou're not doing all 10k at a time, only a subset11:03
fennso 100 oligos per mm^211:04
kanzureyou could easily end up with 1,000 pcr tubes and now you have a different problem...11:05
kanzureeh i guess 1k pcr tubes isn't the end of the world11:05
fennthe m.laboratorium assembly took roughly that many steps11:06
kanzureyashgaroth says, "Honestly I'd be interested just to leverage it for the isothermal amplification + nicking enzyme project from a while back. Suddenly even a 10mer library doesn't sound too insane if you can get 1000x1000 printing resolution, and the ligase assembly process should be somewhat error-correcting. The longer the fragments, the likelier it is to work."11:06
fenni didn't understand that at all11:08
kanzureat one point he mentioned in the logs a nicking enzyme method for dna synthesis and gene assembly.11:08
CaptHindsightdepending on the surface tension figure ~75um dia spots, and maybe a 25um space so maybe 100um center to center11:08
fennhttps://en.wikipedia.org/wiki/Nicking_Enzyme_Amplification_Reaction11:09
nmz787_i1kanzure: but is yash assuming we're printing the 10mers, or do we still need to store and dispense them?11:09
fenn.c 4^1011:11
yoleaux4¹⁰ = 104857611:11
fennwhy do you need 10-mers for nicking enzyme amplification?11:12
nmz787_i1that was a different kind of nicking enzyme trick11:12
nmz787_i1that was from... 2012 I think11:12
-!- nmz787_i1 [~ntmccork@134.134.139.78] has quit [Read error: Connection reset by peer]11:12
-!- eudoxia [~eudoxia@r167-57-2-102.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving]11:13
* fenn looks at http://gnusha.org/logs/2012-02-05.log http://gnusha.org/logs/2012-02-15.log http://diyhpl.us/~bryan/papers2/DNA/nicking-library-method.jpg11:16
fennso this is just to make a crapton of 10mers11:18
fennlike pure white noise11:19
fenni guess each of these beads would be floating in its own microfluidic water bubble? otherwise what would you do with them11:21
-!- Taek42 [~quassel@2001:41d0:1:472e::] has joined ##hplusroadmap11:21
-!- heath_ [~heath@unaffiliated/ybit] has joined ##hplusroadmap11:23
-!- BobaMa_ [bobama@kapsi.fi] has joined ##hplusroadmap11:23
fenni hope they're wrong about having to buy 50 liters of acetonitrile at a time11:26
-!- nsh- [~lol@wikipedia/nsh] has joined ##hplusroadmap11:27
-!- nsh [~lol@wikipedia/nsh] has quit [Disconnected by services]11:28
-!- nsh- is now known as nsh11:28
-!- Netsplit *.net <-> *.split quits: heath, strages, BobaMa, marchtemp, night, Taek, ebowden, Acty, EnabrinTain11:28
-!- Netsplit over, joins: night11:29
-!- nmz787_i [~ntmccork@134.134.139.78] has joined ##hplusroadmap11:30
nmz787_ifenn: the idea was to have a million beads contained individually/separately... then amplify to get a 10-mer, then mix a combo of 10-mers to get your desired sequence11:32
nmz787_ithe nicking amplification was so you only have to synthesize the 10-mers once, then you just amplify any time you need one for your desired sequence11:33
nmz787_iand no, they aren't joking about the diluent needing to be purchased in 50L quantities11:33
nmz787_iwaste management will be a bitch11:33
fenni guess i could see the 6-mer thing working if you had one of those programmable electro-microfluidic arrays11:35
fenncontamination would be annoying though11:36
kanzurei'm having trouble keeping track of the performance of known-methods11:38
fenn1 million individually addressable droplets just seems too hard11:38
kanzurein the microfluidics approach?11:38
fennin any approach, but especially microfluidics11:38
kanzureor in the micropipetting-at-that-resolution-is-too-weird?11:39
kanzure*is-too-weird sense?11:39
fenn1 million is a really big number11:39
fennhave you ever seen 1 million of anything?11:39
kanzuredoes viewing an inkjet printed document count?11:39
kanzureor viewing an lcd screen11:40
fenni think there are something like 100,000 visible stars in the sky11:40
-!- helleshin [~talinck@66-161-138-110.ubr1.dyn.lebanon-oh.fuse.net] has quit [Read error: Connection reset by peer]11:40
kanzurei think another aspect here is that it's all probabilistic anyway11:41
kanzureso.. the only way to get good results is even more quality control :-/ which limits the scale you can do this at anyway.11:41
kanzureunless you have highly parallel quality control11:41
fennhmm you canhttp://i.imgur.com/cl7Ng.jpg11:41
fennyou can only see 5000 stars in the sky apparently11:41
-!- helleshin [~talinck@66-161-138-110.ubr1.dyn.lebanon-oh.fuse.net] has joined ##hplusroadmap11:42
kanzure1k pcr tubes as an output isn't a big deal. i think that would be a good result.11:43
kanzurewe can just manually run gibson on martial-arts-bracketed combinations until something interesting pops out at the end11:43
fenni will smash the plates with my fists until they are really small11:44
kanzurethey should be pre-smashed11:44
fennpyrosequencing can only go up to ~300bp so that should probably be the first assembly size target, which determines everything else11:54
fenn300bp reads11:54
fenni dont really have the whole synthesis sequence assembly protocol laid out straight in my head, but i'm pretty sure it involves pyrosequencing at some point11:55
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap11:56
kanzurethis seems like an okay review, http://diyhpl.us/~bryan/papers2/DNA/Large-scale%20de%20novo%20DNA%20synthesis:%20technologies%20and%20applications%20-%20Church%20-%202014.pdf11:58
-!- nmz787_i [~ntmccork@134.134.139.78] has quit [Quit: Leaving.]12:04
fenneven without doing PCR we will want good temperature control to improve "hybridization stringency" (and get the associated error correction)12:13
fennthis implies something like a pcr tube or pcr plate12:13
fenna laser is not going to cut it12:13
-!- Madplatypus [uid19957@gateway/web/irccloud.com/x-hfvamhwbbeebqzld] has quit [Quit: Connection closed for inactivity]12:16
fennthis is a good paper to read, i recommend anyone participating in this discussion read and try to understand all of it12:18
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 264 seconds]12:18
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap12:41
-!- nmz787_i [~ntmccork@134.134.139.78] has joined ##hplusroadmap12:46
nmz787_ifenn: i've got a few million pixels in front of me right now12:47
nmz787_iseeing them all the time12:47
nmz787_iin my pocket12:47
nmz787_ioh, kanzure said it12:47
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap12:47
-!- ebowden [~ebowden@101.177.244.210] has joined ##hplusroadmap12:47
-!- Acty [uid89656@gateway/web/irccloud.com/x-bkperneclkheqwyj] has joined ##hplusroadmap12:50
-!- EnabrinTain [sid11525@gateway/web/irccloud.com/x-vyhcaaiofprbcbch] has joined ##hplusroadmap12:50
-!- marchtemp [~bitnami@ec2-54-69-125-142.us-west-2.compute.amazonaws.com] has joined ##hplusroadmap12:50
-!- strages [sid11297@gateway/web/irccloud.com/x-lgrxgisyetbdbxnu] has joined ##hplusroadmap12:55
-!- dingo_ is now known as dingo12:56
nmz787_ihttp://hackaday.com/2015/07/14/vintage-vinyl-laser-etched-on-a-tortilla/13:17
nmz787_ikanzure: CaptHindsight here's the old shopping list with some recipe steps at the top https://docs.google.com/spreadsheets/d/1bDsOEVUT5SUXnsP8UowV5iRmUxuk0n_328DqYq_c3J8/edit?usp=sharing13:19
kanzuregevent now has python3 support "officially" (no longer need to use tulipcore) https://github.com/gevent/gevent/issues/38#issuecomment-12100166513:30
kanzuregreenlets and libuv implementation for gevent core loop https://github.com/veegee/guv13:31
-!- Acty [uid89656@gateway/web/irccloud.com/x-bkperneclkheqwyj] has quit [Quit: Connection closed for inactivity]13:51
-!- Acty [uid89656@gateway/web/irccloud.com/x-nnklufjxejowmgxg] has joined ##hplusroadmap14:05
-!- Madplatypus [uid19957@gateway/web/irccloud.com/x-wtoctdqzmnhwtkpi] has joined ##hplusroadmap14:05
kanzurewe should add lots of thermometers everywhere and also some heating elements14:13
kanzurewe could do liquid cooling if necessary14:14
-!- crescendo [~mozart@unaffiliated/crescendo] has quit [Ping timeout: 264 seconds]14:14
kanzurethe los angeles biohackers say that they would be happy to house the machine if necessary, and nmz787 has said yes sorta, so we're good to go i think14:14
-!- crescendo [~mozart@unaffiliated/crescendo] has joined ##hplusroadmap14:16
archelsI didn't read up fully on what you have been talking about, is there a document/wiki page somewhere?14:17
kanzuregimme an email address14:17
juri_do we need a wiki? i can fire one up for this.14:19
kanzure-_____-14:20
kanzurejuri_: did you know that the wiki has been in the /topic for years now?14:20
juri_i meant one for this project specifically. ;)14:20
kanzurearchels: http://diyhpl.us/~bryan/nucleic/dna-inkjet-qt-07121500.pdf14:21
archelskanzure: who is ONE Labs Inc.?14:25
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Ping timeout: 256 seconds]14:26
kanzurearchels: CaptHindsight14:27
kanzurejuri_: you probably have not seen the doc either14:27
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap14:27
-!- eudoxia [~eudoxia@r167-57-2-102.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap14:29
kanzure"modular overlap-directed assembly with linkers (MODAL)" http://openwetware.org/images/c/cd/BricksReview.pdf14:33
-!- nottimschmidt [~timschmid@35.10.220.154] has quit [Ping timeout: 240 seconds]14:36
kanzureyeast that assemble a collection of oligos into the correct order could be selected14:38
kanzureand then for every large-scale plasmid or genome that you print, you also select a bunch of cells that have mutated enzymatic activity that tends to put the oligos together in the right order, toss the colonies that don't.14:38
kanzure</bruteforce>14:39
kanzurehuh the wikipedia article on yeast homologous recombination is.. not good.14:44
juri_kanzure: nope. ;)14:46
-!- eudoxia [~eudoxia@r167-57-2-102.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving]14:47
archelskanzure: sounds cool. What do you want to do with it?14:48
kanzureligase cycling reaction http://aati-us.com/sites/default/files/Amyris_Inc._Rapid_reliable_DNA_assembly_via_ligase_cycling_reaction_AATI_customer_paper.pdf14:48
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-kqmrjiwtgidggfxz] has joined ##hplusroadmap14:48
kanzurearchels: well i'd like to replace about 30 to 50% of all plant dna on the surface of the planet with synthetic alternatives, and get genome snthesis down to $1/genome or lower14:48
kanzurehm that link claims ligase cycling reaction was able to join 20 dna fragments that were each 1 kb?14:49
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-ewaxkskswanwdkwq] has joined ##hplusroadmap15:03
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds]15:13
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap15:22
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection timed out]15:32
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap15:32
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection timed out]15:48
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap15:49
fenncounter culture said yes (see mail)15:58
fennalso note that i am not asleep :\15:59
nmz787_ihow to decide between CCL and L.A.Biohackers?16:06
nmz787_iI think I only know Patrik at CCL... but I know at least Cory and Keoni at L.A.B16:07
justanotheruserDear Justan Otheruser: I’m writing to acknowledge your message and I will write again when I have more information. Love, NCBI Help Desk16:08
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection timed out]16:19
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap16:19
CaptHindsighthmm first no takers now, two16:24
CaptHindsightwe can cut it in half and they can each.....16:24
fennwe can reduce the cost and make ten16:26
fenna dna printer in every garage!16:27
CaptHindsightrecroom16:27
CaptHindsightshould we anodize it to look like wood grain?16:27
superkuhhttps://neurolab.gatech.edu/labs/potter - These guys will sell your email address to spammers.16:29
superkuhThat or they have some serious security issues.16:29
kanzureCaptHindsight: i'll try to get some edits sent to you tonight, sorry for the delay, i'm glad we have some potential homes16:31
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Remote host closed the connection]16:35
kanzurenmz787_i: counter culture labs also has juul (in here) plus ryan bethencourt, patrik dapostraphe, etc.16:36
kanzureParahSailin: i know you hate all of us for not going with the electrode array, but do you have any tricks up your sleeve for gene assembly?16:40
-!- nottimschmidt [~timschmid@c-68-48-117-252.hsd1.mi.comcast.net] has joined ##hplusroadmap16:43
CaptHindsightkanzure: print electrode arrays with it16:44
kanzuregene assembly is weird16:46
Adlaiyou're [a product of] weird16:46
kanzurenottimschmidt: are you familiar with yeast homologous recombination? when attempting to assemble genomes from oligos, we should select yeast that put the oligos in the correct order better. but this unfortunately requires dna sequencing and yeast selection..16:47
kanzure(dna sequencing is to measure the quality of the yeast's work)16:47
kanzure.title http://journals.plos.org/plosone/article?id=10.1371/journal.pone.010732916:56
yoleauxPLOS ONE: A Rapid and Simple Method for DNA Engineering Using Cycled Ligation Assembly16:56
nmz787_ikanzure: a main idea I had was to use expression as a checksum... always make GFP, and discard any transformants that don't glow16:58
nmz787_ior rather, only pick the glowing ones16:58
kanzure"Gibson Assemblies were designed for megabase-sized DNA sequences with hundreds of base pairs of homology, but have been adapted to the kilobase and smaller scale of common laboratory applications such as protein engineering [9]. Gibson Assembly has limitations too, such as increasing inefficiency when the number of inserts increases, inability to assembly small (<100 bp) sequences, and complex and error-prone addition of homologous ...16:59
kanzure... sequences that are needed to direct the orientation and order of the assembled product."16:59
kanzurenmz787_i: yep, agreed.16:59
nmz787_ibbl16:59
nmz787_i/me gone FIBin16:59
kanzure"Thermostable polymerases were the key to PCR, and similarly thermostable ligases [10] have enabled applications based on a cycled-ligation reaction (CLR) [11]. Cycled ligations have been used for a variety of purposes [12]–[15], such as synthesizing genes using synthetic oligonucleotides [16], to directing blunt-end ligation reactions using linker oligonucleotides [17]. Notably, the high specificity of CLRs enables the detection of ...16:59
kanzure... SNPs using a Ligase Chain Reaction (LCR) [18]. Here, we present a DNA fragment assembly method based on a LCR. We use short 40 bp Scaffold Oligonucleotide Connectors (SOCs) that enable directed, scarless, in vitro assembly of multiple DNA fragments into a transformable plasmid in a single reaction. SOCs can be re-used in alternative assembly designs. We apply these cycled ligation assemblies to construct DNA products containing many, ...17:00
kanzure... variably sized, inserts. These cycled ligation assemblies are efficient and easy to design and run."17:00
kanzurei wonder if you have to increase the amount of time you wait for ligase to work as you increase the number of cycles, because otherwise ligase is less likely to bump into the correct location for the larger fragments17:01
kanzure"The molar concentration ratio of SOCs to insert strands proved critical. Too few SOCs in the reaction made early assembly events unlikely, while too many SOCs interfered with assembly in later cycles. With a 5:1 molar ratio, optimal SOCs are needed only to initiate the assembly reaction. Once opposite strands are ligated, they serve as templates for assembling additional complementary strands. In early cycles SOCs were required to ...17:03
kanzure... initiate the assembly of the complementary strands (Figure 1A–D). In later cycles (Figure 1E), if SOCs are bound to already-ligated strands they compete with productive assembly of complementary strands. The 5:1 ratio (Figure 1F, lanes 4) is the optimal balance between these conflicting requirements."17:03
-!- nmz787_i [~ntmccork@134.134.139.78] has quit [Read error: Connection reset by peer]17:04
kanzure"Easily designed Scaffold Oligonucleotide Connectors (containing the last 20 bp of the first sequence and the first 20 bp of the next) guide and direct the assembly, but are not incorporated into the final product. The same amplified starting fragment can be reused from one DNA assembly plan to the next, with only the SOCs changing to guide alternative constructions. A given sequence (such as for GFP) need only be amplified once to be ...17:06
kanzure... assembled into different locations in any number of constructs. This is in stark contrast to comparable assembly strategies, where each insert must encode its own assembly instructions through attached restriction sites or homologous sequences, hampering the reusability of a sequence from one assembly reaction to the next."17:06
kanzure"The specificity of CLAs even allows multiple distinct assembly reactions to take place simultaneously in the same tube. Inclusion of a vector backbone in a CLA enables the direct transformation of assembled circular constructs into competent cells immediately following assembly. However, CLAs are not ideal in all situations. Due to cycles of denaturation and annealing, highly homologous insert sequences can interfere with each other’s ...17:07
kanzure... assembly, reducing efficiency. Gibson assemblies share this limitation."17:07
kanzureok, well it sounds like this method would benefit from inserting some "junk dna" once in a while to attempt to prevent highly homologous sequences17:08
kanzureor maybe using unnatural nucleotides could help here.17:08
kanzure"The efficiency of CLAs decreases as the size of the assembled DNA fragments grows above 5–6 kb. This effect is likely due to the probabilistic nature of CLAs, where productive assembly is due to four-part interactions between two single stranded DNA fragments, a SOC, and Taq ligase. As the length of DNA fragments increases, these rare events are crowded out by the higher probability of off-target binding between various combinations ...17:09
kanzure... of SOCs and DNA fragments."17:09
kanzurewell, increase the time then.... duh?17:09
fennwhat does a "cycle" consist of?17:09
kanzure"At the scale of the most common DNA assembly goals, using fragments from <500 bp to 5 kb, cycled-ligation assemblies are efficient and powerful, able to assemble many inserts in a single reaction."17:09
kanzurefigure 1a17:10
kanzurefigure 1a shows a cycle17:10
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-ewaxkskswanwdkwq] has quit [Quit: Connection closed for inactivity]17:10
fennso its just raising the temperature to 95C, then letting it cool to 60C to be ligated17:13
kanzurealso in addition to increasing the time or duration during the later cycles, you could also decrease the volume of the liquid by various forms of trickery, which should increase the probability of the correct events happening17:14
fenni guess i figured this was how it worked and why would anyone do it differently17:16
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 250 seconds]17:16
kanzurei think most people synthesize some stuff at the start and end of each strand, and then they hope polymerase will match them up17:17
fennthere was something about "chewing back" to expose matching ends17:17
fennbut that wouldn't leave a scar17:17
kanzurethe downside of cycled ligation assembly is that all of the oligos in all the spots need to be converted to dsDNA first, and then also there needs to be some printed extra oligos to serve as the connectors17:18
fennyeah you don't end up saving any base pairs because you have to print the connectors too (if saving on base pairs were the point, which it probably isn't)17:18
fenni don't think it needs to be converted to dsDNA first17:19
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap17:20
fennmaybe that's the point of this method; it doesn't need to be dsDNA17:21
kanzure"A thermostable ligase (green) specific for nicks in dsDNA seamlessly ligates the two bottom strands. (E) Subsequent cycles of denaturation and annealing/ligation allow the thermostable ligase to ligate the top strands, using the ligated bottom strands as a template. Further cycles of ligation exponentially increase the amount of correct product as the products of early rounds served as an expanding pool of template for later cycles."17:22
fennthe connector makes it dsDNA for like 40bp17:22
fennanyway you want to make dsDNA from your oligos so you can reject mismatches due to errors17:23
fenntrying to cheap out and overlap only the minimum lets errors sneak through in the parts with no overlaps17:24
kanzureright, right, bolting on some quality control to the original synthesis would be nice as well17:27
fennthat's what i was talking about17:27
kanzuremaybe some fluorophores can be purchased attached to the phosphoramidites or something17:28
kanzurecory just suggested the same "place a drop of water/solvent near the spots that you want to combine, then grab via pipette" idea a few moments ago.17:34
kanzurehe also suggests using a photocleavable linker and using a laser or dmd to select which spots of interest you want to pick up17:34
fennyou'd want orange-colored plexiglass in the enclosure then17:35
kanzureso we still don't have imaginary number wavelength lasers?17:37
kanzurewhere's our physicist17:37
fennwe do, but the light they emit travels a right angles to reality17:37
fennthe "place a drop of water" is equivalent to "wall-less wells" based on surface patterning/coating of the glass slide with rain-x17:38
kanzureright17:39
fennthis is good because it lets you keep droplets separate but you don't have walls in the way of the print head17:40
fennhowever it doesn't let you do PCR right on the plate, is this a problem?17:40
fennstupid water17:41
kanzurewhich one doesn't let you do pcr?17:41
fenns/PCR/stringent hybridization/17:41
fennhaving no walls17:41
fennthe water would just evaporate from the droplet at 95C17:42
kanzureah right this is why the other stuff was in an oil bath17:42
kanzurehmph17:42
kanzurewell look- at some point, the output matter does have to get off of the surface and into at least 1 tube17:42
fennharrumph17:43
kanzurewas the goal to avoid pipetting tiny drops and tiny pores?17:44
fennyou want to do stringent hybridization as early as possible17:44
fennthe goal was to reduce the amount of pipetting17:45
kanzurefor some reason i keep forgetting that gel extraction is a thing that is supposed to work17:45
fennit's a pain though17:45
fennwhy the hell do people use gels17:45
fennthere should be a commonly available inexpensive DNA length spectrometer17:46
fennjust a chromatography column based on gel electrophoresis17:46
fennthen at the end you have a UV LED and photodiode to detect the dna coming out17:47
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap17:47
fennViper168: hey are you the laser guy17:47
CaptHindsighta laser to photocleave?17:49
-!- Douhet [~Douhet@unaffiliated/douhet] has quit [Ping timeout: 246 seconds]17:49
-!- Douhet [~Douhet@unaffiliated/douhet] has joined ##hplusroadmap17:49
CaptHindsightwhat spot size and what wavelength does it require?17:50
kanzurei don't understand "you don't have walls in the way of the print head"17:50
kanzureCaptHindsight: still deciding if it's necessary17:50
fennthe drops exit the print head at some angle close to 90 degrees but not exactly 90 degrees... the farther away from the surface you are, the more position error you get in where the drops land17:52
fennso the head should be as close as possible to the surface for best accuracy17:53
CaptHindsightit'll be around 1mm17:53
CaptHindsightthe drops also need to form in flight17:54
CaptHindsightit also makes a difference if the head is stationary or moving17:54
kanzurethey wont be drops unless it's liquid- the default azco phosphoramidites are solid17:55
fennits a brown bottle with 1 gram of solid in it?17:55
kanzurehrmmm17:55
CaptHindsightwhat vehicle is used in the liquid versions?17:56
fennit might make sense if the goal is to prevent water contamination17:56
CaptHindsightsolid block, powder, goo?17:56
fennvehicle?17:57
fenni assumed the solvent was acetonitrile17:57
CaptHindsightsolvent/vehicle, liquid carrier17:57
fennthinking about running this thing... if it just dumps a stream of acetonitrile on the plate with each pass, that's going to create a lot of waste solvent18:01
fennit seems like a fog gun or high pressure sprayer would use much less solvent18:01
-!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has joined ##hplusroadmap18:02
fennbut it's a sealed box so you have to draw "air" from inside the box18:02
CaptHindsight3-methoxypropionitrile18:05
CaptHindsight2-methyl glutaronitrile18:05
fennwhy is it different18:05
fennacetonitrile has relatively low toxicity compared to i.e. propionitrile18:06
CaptHindsighthave to see how much solvent wash is really needed18:07
fennhow did posam blow-dry the slides even though it was a sealed container?18:08
CaptHindsightthe drying nozzles are next to the printhead18:09
fennbut what kept the enclosure from turning into a balloon18:10
CaptHindsightand they also scrub the gas inside for moisture18:10
CaptHindsight"The reagents are currently removed from the slides by an inert gas stream. Increasing the size of the stream or replacing it with a differentmechanisism should be investigated"18:12
CaptHindsightfenn: positive pressure inside the enclosure with a check valve to outside atmosphere18:16
CaptHindsightthe POSaM manuals are not the best write up, but certainly not the worst18:18
CaptHindsightwish they would have included more pics or had a demo video of operation18:19
CaptHindsightI could have skipped trying to read their minds in certain sections18:19
kanzurefenn: juul: heath says he is at CCL staring awkwardly at three other people that don't seem to be fenn or juul18:19
kanzureCaptHindsight: posam was built before the invention of youtube. ouch.18:19
CaptHindsightand thankfully twitter18:20
fennlol ok i guess i will head over to ccl then18:20
fenneta 35 minutes18:20
-!- Burn_ [~Burn@pool-71-241-254-153.washdc.fios.verizon.net] has quit [Disconnected by services]18:21
kanzurefenn: let me check with him first18:21
CaptHindsighthasta18:21
CaptHindsightgoing back to reading People18:21
-!- Burnin8 [~Burn@pool-71-241-254-153.washdc.fios.verizon.net] has joined ##hplusroadmap18:21
kanzureis that how you use your down time? :-)18:21
CaptHindsightdown time for me is usually passing out18:22
kanzurefenn: whoops i don't know a current phone number for him. i sent him an email (and cc'd you) instead.18:23
fennold? 256 274 4225 new? 347 430 740618:27
kanzureold: 256 274 4225 and 256 740 203718:30
kanzureno answer on 347. whatever.18:34
ParahSailindunno18:54
kanzureParahSailin: opinion on the ligase protocol in http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0107329 ?18:58
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 252 seconds]19:04
ParahSailinligase doesnt usually misbehave19:10
kanzure.wik dna ligase19:15
yoleaux"In molecular biology, DNA ligase is a specific type of enzyme, a ligase, (EC 6.5.1.1) that facilitates the joining of DNA strands together by catalyzing the formation of a phosphodiester bond." — https://en.wikipedia.org/wiki/Dna_Ligase19:15
ParahSailinat least you're trying something19:17
kanzurehm?19:20
kanzure.wik ligation (molecular biology)19:21
yoleaux"In molecular biology, ligation is the joining of two nucleic acid fragments through the action of an enzyme. It is an essential laboratory procedure in the molecular cloning of DNA whereby DNA fragments are joined together to create recombinant DNA molecules, such as when a foreign DNA fragment is inserted into a plasmid." — https://en.wikipedia.org/wiki/Ligation_(molecular_biology)19:21
-!- delinquentme [~delinquen@74.61.157.78] has quit [Ping timeout: 265 seconds]19:36
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap19:41
-!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has joined ##hplusroadmap19:45
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: Namaste]20:15
fennno sign of heath here20:20
kanzureoh i thought i had stopped you in time20:40
kanzure"students who try a homology-based gene assembly method have a high probability of success on the first try" http://www.biomedcentral.com/content/pdf/s13036-015-0006-z.pdf21:00
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 264 seconds]21:07
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap21:31
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 246 seconds]21:33
-!- ebowden [~ebowden@101.177.244.210] has quit [Remote host closed the connection]21:37
-!- ebowden [~ebowden@101.177.244.210] has joined ##hplusroadmap21:38
-!- Acty [uid89656@gateway/web/irccloud.com/x-nnklufjxejowmgxg] has quit [Quit: Connection closed for inactivity]21:41
-!- Porb [~Porbus@CPE-124-176-219-129.lns5.win.bigpond.net.au] has joined ##hplusroadmap22:05
-!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has quit [Ping timeout: 244 seconds]22:38
-!- nmz787_i [~ntmccork@192.55.55.41] has joined ##hplusroadmap22:40
nmz787_ifenn: they likely invented chemical synthesis before finding/mutating a thermostable ligase22:49
nmz787_iso the FIB wizard said if I make some trenches in bulk silicon that backed up next to each other, then turn the wafer sideways to pierce through the wall separating the two, that he could probably make a 60nm hole. Piercing through gold leaf or some other metal leaf would be easier and he could get a lot smaller (because it's thinner, so less redeposition)22:51
nmz787_ibut that the leaf would disintegrate with surface tension most likely22:51
nmz787_ianother idea I had was making he same trenches, but then connecting the two with a nano trench... such that it acted the same as a pore/channel with a cover-slip on top22:52
nmz787_ibut the wizard said that he thought getting the cover slip on top would be troublesome22:53
nmz787_ikanzure: that ligase technique is essentially what I was thinking, except I thought to make the SOCs the same length as the other fragments, since the starting oligos wouldn't be dsDNA anyway.22:56
nmz787_iwell, I guess I also thought that using the adapters up would be a good thing as far as keeping the reaction zone tidy22:57
nmz787_ifenn: they used different solvent than acetonitrile because they said it made efficiency increases22:57
-!- knobuddy is now known as TheButler23:28
-!- proofoflogic [uid65184@gateway/web/irccloud.com/x-nmhzmfuffheoxtss] has joined ##hplusroadmap23:41
-!- Madplatypus is now known as BearyMad23:45
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds]23:50
-!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap23:51
-!- Viper168_ is now known as Viper16823:56
--- Log closed Wed Jul 15 00:00:11 2015

Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!