--- Log opened Fri Apr 13 00:00:47 2018
00:09 -!- TMA [tma@twin.jikos.cz] has quit [Ping timeout: 240 seconds]
00:10 < archels> kanzure: where the heck am I going to leave it, I don't even have a car
00:10 < archels> in my personal experience, even hackerspaces aren't too happy if you dump equipment in their backyard
00:10 < archels> (even cool equipment)
00:11 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 240 seconds]
00:13 -!- TMA [tma@twin.jikos.cz] has joined ##hplusroadmap
00:13 -!- archels [charl@toad.stack.nl] has quit [Changing host]
00:13 -!- archels [charl@unaffiliated/archels] has joined ##hplusroadmap
00:25 -!- wrldpc1 [~ben@hcccbcac120.bai.ne.jp] has quit [Quit: wrldpc1]
00:26 -!- jtimon [~quassel@142.29.134.37.dynamic.jazztel.es] has quit [Ping timeout: 256 seconds]
00:32 -!- aeiousomething [~aeiousome@124.123.14.99] has quit [Ping timeout: 265 seconds]
00:53 -!- wrldpc1 [~ben@softbank126225109178.bbtec.net] has joined ##hplusroadmap
01:02 < nmz787> archels: what city?
01:02 < nmz787> archels: any idea what the equipment actually was?
01:02  * nmz787 has car and can drive it well over long distances
01:53 < nmz787> http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005268
01:53 < nmz787> .title
01:53 < yoleaux> Could a Neuroscientist Understand a Microprocessor?
01:53 < nmz787> "perhaps there is a Donkey Kong transistor or a Space Invaders transistor. Even if we can lesion each individual transistor, we do not get much closer to an understanding of how the processor really works."
01:57 -!- jtimon [~quassel@142.29.134.37.dynamic.jazztel.es] has joined ##hplusroadmap
02:00 < ebowden> lol
02:05 < nmz787> superkuh: http://atlc.sourceforge.net/
02:05 < nmz787> CaptHindsight: ^
02:06 < nmz787> .tw https://twitter.com/bunniestudios/status/984396916514369536
02:06 < yoleaux> Q: What's the effect of putting a finger (or water) on an edge coupled microstrip diff pair. A: Drops impedance from Zdiff=90 to 62 ohms. Intuition: hi-K of water "smushes" E-field into PCB, making the pair "feel" ground more, lowering Z. Thx to http://atlc.sourceforge.net/ #TMYK 💫 https://pbs.twimg.com/media/DalFJyoUQAEnFzv.jpg https://pbs.twimg.com/media/DalFK9BUQAAficw.jpg (@bunniestudios)
02:06 -!- darsie [~username@84-114-73-160.cable.dynamic.surfer.at] has joined ##hplusroadmap
02:11 < ebowden> nmz787, you interested in developing your own tiny chip fab like that other guy did?
02:19 < archels> nmz787: I'm in The Netherlands
02:20 < archels> I only know the DIYbio people in De Waag (Amsterdam) who might be interested in this kind of thing
02:41 -!- jtimon [~quassel@142.29.134.37.dynamic.jazztel.es] has quit [Ping timeout: 260 seconds]
02:44 < archels> incidentally, this is a reasonaly cogent reply to that "could a neuroscientist understand a microprocessor" paper: https://www.sciencedirect.com/science/article/pii/S0896627316310406
02:45 < archels> although an alternative cogent reply is to call the whole comparison stupid from the outset. obviously you would use different approaches and develop different techniques to reverse engineer a microprocessor than you would for a brain
02:47 < archels> nmz787: but thanks for the offer :D
03:21 -!- bluebear_ [~dluhos@80.95.97.194] has joined ##hplusroadmap
03:48 -!- gabbar1947 [uid205515@gateway/web/irccloud.com/x-xccnbophqmmwvfka] has joined ##hplusroadmap
03:53 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
04:23 -!- Gurkenglas [~Gurkengla@dslb-188-102-077-078.188.102.pools.vodafone-ip.de] has joined ##hplusroadmap
04:32 < superkuh> nmz787, cool. I'll have to try that out.
04:58 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 276 seconds]
04:58 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
05:18 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has joined ##hplusroadmap
05:26 -!- aeiousomething [~aeiousome@124.123.14.99] has joined ##hplusroadmap
05:28 -!- jtimon [~quassel@142.29.134.37.dynamic.jazztel.es] has joined ##hplusroadmap
05:41 -!- ebowden [~ebowden@128.250.0.209] has quit [Remote host closed the connection]
05:42 -!- ebowden [~ebowden@128.250.0.209] has joined ##hplusroadmap
05:46 -!- aeiousomething [~aeiousome@124.123.14.99] has quit [Ping timeout: 265 seconds]
05:46 -!- ebowden [~ebowden@128.250.0.209] has quit [Ping timeout: 264 seconds]
05:59 < kanzure> archels: rent a car.
06:03 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 255 seconds]
06:03 -!- preview_ [~quassel@203.167.243.242] has joined ##hplusroadmap
06:16 < kanzure> .title https://news.ycombinator.com/item?id=16827248
06:16 < yoleaux> “Is curing patients a sustainable business model?” Goldman Sachs analysts ask | Hacker News
06:17 -!- aeiousomething [~aeiousome@124.123.14.99] has joined ##hplusroadmap
06:19 < kanzure> i wonder if anyone has published the alternative transhumanist pro-health pro-upgrade view (sell upgrades to people -> they become more productive -> they have more money to spend on your services).
06:36 -!- aeiousomething [~aeiousome@124.123.14.99] has quit [Ping timeout: 265 seconds]
06:48 < archels> kanzure: now you have me picturing myself as a street vendor selling illegal biotech equipment out of the back of a van. "psssst. hey. want to buy some DNA synthesizers?"
07:07 -!- preview_ [~quassel@203.167.243.242] has quit [Ping timeout: 245 seconds]
07:07 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
07:21 < kanzure> archels: say yes to life.
07:22 < kanzure> "Rewritable multi-event analog recording in bacterial and mammalian cells" http://science.sciencemag.org/content/360/6385/eaap8992
07:24 < kanzure> ehhh not so convinced about this technique-- "signals are recorded in the form of plasmid ratios"
07:25 < kanzure> "We demonstrate that the ratio of the recording plasmid pair in CAMERA 1 can be stably maintained in bacteria over 144 hours and a dilution ratio of 1017. By using a writing complex of the Cas9 nuclease and a guide RNA to selectively target one of the recording plasmids, we can cause this plasmid ratio to shift in a dose-dependent manner. The presence or absence of a stimulus is recorded in ...
07:25 < kanzure> ...CAMERA 1 by linking to the expression of the writing complex. The analog format of CAMERA 1 enables recording of signal amplitude over a known time scale, or recording of the duration of a signal of known strength. Two resetting methods enable cells harboring CAMERA 1 to function over repeated cycles of recording and erasing."
07:25 < kanzure> "CAMERA 2 uses base editors to record stimuli of interest as permanent single-base modifications in cellular DNA. Predictable and dose-dependent accumulation of base editing was observed over 68 generations in bacteria. CAMERA 2 achieved analog recording of multiple stimuli of interest, including exposure to antibiotics, nutrients, viruses, and light. When recording to a high-copy plasmid, ...
07:25 < kanzure> ...CAMERA 2 provides reliable readout by sequencing only 10 to 100 cells and can record event order using an overlapping guide RNA design.
07:25 < kanzure> "
07:25 < kanzure> multi-generational recording is interesting tho
07:25 < kanzure> wiht just plasmid ratios, i don't think you can record sequences of events?
07:34 -!- gabbar1947 [uid205515@gateway/web/irccloud.com/x-xccnbophqmmwvfka] has quit [Quit: Connection closed for inactivity]
07:35 < kanzure> is there any non-mRNA ribosome activity that biologists have overlooked because central dogma?
07:42 < kanzure> can we get something like "rolling circle amplification" to work with circular mRNA plasmid and a ribosome?
07:43 < kanzure> .wik ribosome flow model on a ring
07:43 < yoleaux> kanzure: Sorry, that command (.wik) crashed.
07:44 < kanzure> "Through simultaneous interactions with the cap-binding protein eIF4E and the poly(A)-binding protein PABP, the eukaryotic initiation factor eIF4G is able to bridge the two ends of the mRNA[21, 22]. This suggests that a large fraction of the ribosomes that complete translating the mRNA re-initiate. The RFMR is a good approximation of the translation dynamics in these circularized mRNAs. In ...
07:44 < kanzure> ...addition, circular RNA forms (which includes covalent RNA interactions) appear in all domains of life[23,24,25,26,27,28,29,30], and it was recently suggested that circular RNAs can be translated in eukaryotes[28,29,30]."
07:44 < kanzure> from a modeling study https://www.nature.com/articles/s41598-017-09602-6
07:45 < kanzure> [28] Abe, N. et al. Rolling circle translation of circular RNA in living human cells. Sci. Rep. 5, 16435 (2015).
07:45 < kanzure> [29] Granados-Riveron, J. T. & Aquino-Jarquin, G. The complexity of the translation ability of circRNAs. Biochimica et Biophysica Acta (BBA) - Gene Regulatory Mechanisms 1859, 1245–1251 (2016).
07:45 < kanzure> [30] AbouHaidar, M., Venkataraman, S., Golshani, A., Liu, B. & Ahmad, T. Novel coding, translation, and gene expression of a replicating covalently closed circular RNA of 220 nt. Proc. Natl. Acad. Sci. USA 111, 14542–14547 (2014).
07:46 -!- augur [~augur@2600:380:872c:8b74:446:5d4a:54f8:ebeb] has joined ##hplusroadmap
07:47 < kanzure> "Novel coding, translation, and gene expression of a replicating covalently closed circular RNA of 220 nt" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4209996/
07:47 < kanzure> "The highly structured (64% GC) covalently closed circular (CCC) RNA (220 nt) of the virusoid associated with rice yellow mottle virus codes for a 16-kDa highly basic protein using novel modalities for coding, translation, and gene expression. This CCC RNA is the smallest among all known viroids and virusoids and the only one that codes proteins. Its sequence possesses an internal ribosome ...
07:47 < kanzure> ...entry site and is directly translated through two (or three) completely overlapping ORFs (shifting to a new reading frame at the end of each round) ... This circular RNA with no noncoding sequences is a unique natural supercompact “nanogenome.”"
07:49 < kanzure> "Rolling circle translation of circular RNA in living human cells" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4639774/
07:49 < kanzure> "... In this study, we prepared circular RNAs with an infinite open reading frame and tested their translation in eukaryotic systems. Circular RNAs were translated into long proteins in rabbit reticulocyte lysate in the absence of any particular element for internal ribosome entry, a poly-A tail, or a cap structure. The translation systems in eukaryote can accept much simpler RNA as a ...
07:49 < kanzure> ...template for protein synthesis by cyclisation. Here, we demonstrated that the circular RNA is efficiently translated in living human cells to produce abundant protein product by RCA mechanism."
07:50 < kanzure> uh... this is extremely useful.
07:51 -!- ebowden [~ebowden@128.250.0.209] has joined ##hplusroadmap
07:52 < kanzure> p. sure that ribosomal transient mutators and mistranslation agents are well studied.
07:52 < kanzure> seeing as how that's how many antibiotics function, and such.
07:56 < kanzure> "The frequency of translational misreading errors in E. coli is largely determined by tRNA competition" http://rnajournal.cshlp.org/content/13/1/87.short
08:12 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 260 seconds]
08:12 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
08:52 -!- Gurkenglas [~Gurkengla@dslb-188-102-077-078.188.102.pools.vodafone-ip.de] has quit [Ping timeout: 240 seconds]
08:57 < kanzure> also: the problems related to high drug/pharma prices can probably be resolved if drug development was extremely cheap instead of being extremely expensive
09:15 -!- catern [~catern@catern.com] has quit [Quit: catern]
09:17 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 256 seconds]
09:17 -!- preview_ [~quassel@203.167.243.242] has joined ##hplusroadmap
09:21 < kanzure> "Structural basis for the inhibition of the eukaryotic ribosome" http://pubman.mpdl.mpg.de/pubman/item/escidoc:2060828/component/escidoc:2060829/2060828.pdf
09:27 -!- augur [~augur@2600:380:872c:8b74:446:5d4a:54f8:ebeb] has quit [Remote host closed the connection]
09:33 -!- aeiousomething [~aeiousome@124.123.14.99] has joined ##hplusroadmap
09:33 -!- aeiousomething [~aeiousome@124.123.14.99] has quit [Client Quit]
09:34 -!- aeiousomething [~aeiousome@124.123.14.99] has joined ##hplusroadmap
09:40 -!- emeraldgreen [~user@188.170.80.112] has joined ##hplusroadmap
09:44 -!- catern [~catern@catern.com] has joined ##hplusroadmap
10:02 -!- emeraldgreen [~user@188.170.80.112] has quit [Ping timeout: 276 seconds]
10:08 -!- NikopolSohru [~NSohru@185.212.171.198] has joined ##hplusroadmap
10:21 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
10:21 -!- preview_ [~quassel@203.167.243.242] has quit [Ping timeout: 260 seconds]
10:30 -!- l_wl [~l_wl@pool-173-66-205-23.washdc.fios.verizon.net] has joined ##hplusroadmap
10:33 < kanzure> .title https://synbiobeta.com/darpa-awards-ginkgo-bioworks-and-transcriptic/
10:33 < yoleaux> DARPA Awards Ginkgo Bioworks and Transcriptic $9.5M to Bring AI into the Lab - SynBioBeta
10:34 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-wkykjngfzgbkfcbe] has joined ##hplusroadmap
11:18 -!- l_wl [~l_wl@pool-173-66-205-23.washdc.fios.verizon.net] has quit [Remote host closed the connection]
11:18 -!- l_wl [~l_wl@pool-173-66-205-23.washdc.fios.verizon.net] has joined ##hplusroadmap
11:26 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 276 seconds]
11:26 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
11:33 -!- bluebear_ [~dluhos@80.95.97.194] has quit [Quit: Leaving.]
11:37 < nmz787> huh, transcriptic is still around eh?
11:37 < nmz787> good for them
11:39 < nmz787> kanzure: that virus is into demoscene methinks
11:39 < nmz787> or maybe demoscene is into RCA virii?
11:42 -!- sandeepkr [~sandeepkr@ec2-52-29-251-54.eu-central-1.compute.amazonaws.com] has quit [Ping timeout: 256 seconds]
11:43 < nmz787> ebowden: the answer to that depends. I'd rather not try to wear too many hats... I know more about fab (which is a lot, but not intermediary or expert knowledge at all) than I do about transistor and other metamaterial design
11:44 < ebowden> Ah, ok.
11:44 -!- sandeepkr [~sandeepkr@ec2-52-29-251-54.eu-central-1.compute.amazonaws.com] has joined ##hplusroadmap
11:44 < nmz787> so basically I need another few hundred years if I really wanted to DIY everything
11:44 < ebowden> lol
11:44 < nmz787> yeaaahh
11:45 < ebowden> Well, at a larger process node, you could still produce a bunch of chips that have gone out of production.
11:45 < nmz787> lately I'm thinking I'll try decapping existing ICs and mate them with micro/nanofluidics to achieve proof-of-concept
11:45 < nmz787> (and maybe something useful?)
11:45 < ebowden> (8bitguy would love you forever.)
11:45 < nmz787> at this point I need to focus on clearing my project backlog, versus adding to it (at least in earnest, I can always take notes)
11:46 < nmz787> I am getting ready to take a (paid) month off of work
11:46 < nmz787> and have another one of those in backup for an immediate extension or to use sometime in next 12 months
11:47 < nmz787> (job policy for when you have a kid)
11:47 < nmz787> haha but kid is another project I guess... but at least I'll be home and able to read shit while it sleeps
11:47 < nmz787> or tinker
11:47 < ebowden> Oh, what do you do for work?
11:47 < nmz787> software at Intel
11:50 < CaptHindsight> nmz787: sleep, what is this sleep with a newborn you speak of?
11:50 < nmz787> CaptHindsight: I heard they have a 1 minute on/ 14 minute off duty cycle :)
11:50 < CaptHindsight> hah
11:51 < nmz787> at least in the beginning
11:51 < CaptHindsight> you'll catch up on sleep in 18-24 months maybe
11:51 < nmz787> I am a heavy sleeper, challenge accepted
11:52  * nmz787 is still in denial stage of acceptance
11:52 < CaptHindsight> nobody warned us, sleep as much as you can before
11:54 < nmz787> I assume you're talking about the actual delivery day?
11:54 < CaptHindsight> what's is the tiniest microcontroller die?
11:54 < CaptHindsight> nmz787: yes, before that, after is too late  :)
11:55 < nmz787> CaptHindsight: modern processors have several microcontrollers inside
11:55 < nmz787> so, uh, I'd guess sub-mm^2
11:56 < CaptHindsight> yeah, but are they available in die form
11:57 < nmz787> ah, you'd need to search digikey I guess
11:57 < CaptHindsight> single micro several microns square
11:58 < ebowden> CaptHindsight, I am curious, why are you searching for tiny microcontrollers?
11:59 < CaptHindsight> ebowden: wondering if anyone has actually fabbed any
11:59 < CaptHindsight> down to cell size
11:59 < ebowden> You want to put one inside a cell?
11:59 < CaptHindsight> sure
12:00 < ebowden> Whatever for?
12:04 < CaptHindsight> at 14nm you could fit a 8b micro only a few um^2
12:04 < CaptHindsight> ebowden: initially for sensors
12:05 < kanzure> "Intracellular silicon chips in living cells" http://diyhpl.us/~bryan/papers2/bio/Intracellular%20silicon%20chips%20in%20living%20cells%20-%202010.pdf
12:07 < CaptHindsight> also needs a glucose battery
12:09 < CaptHindsight> Strategic Analysis, Inc  and Booz Allen Hamilton are dabbling with the DNA memory
12:10 < CaptHindsight> why pass up free money
12:19 < nmz787> CaptHindsight: I can send a ref on inside-cell sized electronics... reviewed an NSF grant proposal on that topic a few months ago
12:23 < nmz787> "current transistors have ~14-28nm critical feature sizes, which allows for ~50 transistors in ` um^2, enough to make useful circuits"
12:24 < adlai> archels: read neuromancer
12:25 < adlai> gradstudentbot: resurrect thyself
12:25 < adlai> paperbot: you too
12:25 < nmz787> CaptHindsight: actually I guess there's nothing public to share
12:28 -!- wrldpc1 [~ben@softbank126225109178.bbtec.net] has quit [Ping timeout: 264 seconds]
12:29 < CaptHindsight> nmz787: I don't want to reinvent the wheel is why I was looking
12:30 -!- preview_ [~quassel@203.167.243.242] has joined ##hplusroadmap
12:31 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 268 seconds]
12:31 < kanzure> just use bigger cells if you want to insert chips
12:32 < kanzure> there's lots of huge cells out there
12:32 -!- hazirafel [~hazirafel@bzq-109-66-0-197.red.bezeqint.net] has joined ##hplusroadmap
12:32 < CaptHindsight> nature decides the cell and their sizes
12:32 -!- wrldpc1 [~ben@softbank126225109178.bbtec.net] has joined ##hplusroadmap
12:34 < CaptHindsight> few um dia for a chip, battery and membrane would be the first real goal
12:34 < CaptHindsight> red blood cell size
12:39 -!- drewbot [~cinch@54.166.94.209] has joined ##hplusroadmap
12:44 < ebowden> Maybe an ATP battery.
12:45 -!- ebowden [~ebowden@128.250.0.209] has quit [Remote host closed the connection]
12:46 -!- ebowden [~ebowden@128.250.0.209] has joined ##hplusroadmap
12:47 < kanzure> can you give a ref for microchip use of ATP as energy source..?
12:59 < CaptHindsight> https://en.wikipedia.org/wiki/Biobattery
12:59 < CaptHindsight>  E. coli battery
13:04 < nmz787> CaptHindsight: other guy was working towards radio-wave capture for energy
13:05 < CaptHindsight> yeah, like RFID
13:06 < CaptHindsight> but glucose is usually in most bloodstreams
13:21 < CaptHindsight> Using bacteria batteries to make electricity  https://www.sciencedaily.com/releases/2013/07/130717051733.htm
13:21 < CaptHindsight> http://ekvv.uni-bielefeld.de/blog/uninews/entry/using_bacteria_batteries_to_make
13:26 < CaptHindsight> maybe clockless microcontroller made of neurons
13:28 < CaptHindsight> grown in 3d to take up less space
13:34 -!- preview_ [~quassel@203.167.243.242] has quit [Ping timeout: 256 seconds]
13:34 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
13:36 < CaptHindsight> what organism has the smallest brain? (hows that for a comedic setup?)
13:39 < CaptHindsight> http://www.bionet.ee.columbia.edu/hackathons/ffbh/2017
13:49 < kanzure> depends on what you think a brain is. if you count by number of neurons then there's some womrs with 301 neurons.
14:06 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 264 seconds]
14:22 < CaptHindsight> https://www.sciencealert.com/scientists-put-worm-brain-in-lego-robot-openworm-connectome
14:22 < CaptHindsight> with video
14:23 < CaptHindsight> enough to run a simple robot or work a TV remote
14:32 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
14:39 -!- hazirafel [~hazirafel@bzq-109-66-0-197.red.bezeqint.net] has quit [Ping timeout: 240 seconds]
14:43 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 240 seconds]
15:10 < fenn> with the endless loop of RNA being translated, do you get a big single protein chain or multiple proteins?
15:14 -!- NikopolSohru [~NSohru@185.212.171.198] has quit [Ping timeout: 264 seconds]
15:17 < fenn> i read about a robot using a microbial fuel cell that hunted slugs with machine vision and then "ate" them by putting the dead slugs into its fuel cell
15:17 < fenn> it was a very slow robot but energy independent
15:18 < kanzure> with endless circularized mRNA translation it looks like it can do both single protein chain and also multiple, depending on which codons you use.
15:19 < fenn> do you use a stop codon to finish? how does it end?
15:21 < kanzure> this one has endless translation with no stop codon https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4639774/
15:21 < kanzure> "long-repeating peptide sequence" in this case
15:23 -!- CandleGlow [~CandleGlo@unaffiliated/candleglow] has joined ##hplusroadmap
15:27 -!- aeiousomething [~aeiousome@124.123.14.99] has quit [Quit: goodnight!]
15:28 < kanzure> this one seems to have a termination sequence https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4209996/
15:58 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-wkykjngfzgbkfcbe] has quit [Quit: Connection closed for inactivity]
16:05 < fenn> looks a little more complicated than that. sometimes it terminates, sometimes it doesn't, with the same sequence
16:06 < fenn> anyway it answers my question, which is whether the loop falls off the ribosome when there's a termination codon
16:06 < fenn> it doesn't fall off
16:07 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
16:12 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has quit [Ping timeout: 240 seconds]
16:13 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has joined ##hplusroadmap
16:16 < kanzure> i want this for molecular nanotechnology reasons
16:17 < kanzure> you have a molecular assembler right there
16:17 < kanzure> (if you had stepwise ratcheting or incorporation override)
16:17 < nmz787> seems like number of nucleotides matter, i.e. if you can n%3 it
16:17 < nmz787> then you can get some interesting frameshift stuff going on
16:18 < nmz787> then top that with we dont knot protein folding... secondary/tertiary effects probably can come into play... i.e. self-catalytic stuff that only works during setup phase (i.e. it breaks itself, or hides the catalytic domain inside the protein core)
16:18 < nmz787> s/dont knot/don't know/
16:19 < nmz787> catalytic which could be cut, cut+splice, fuse... all the above
16:19 -!- aeiousomething [~aeiousome@124.123.14.99] has joined ##hplusroadmap
16:20 < nmz787> do any of the recent computational reverse-synthesis (retrosynthesis) articles publish code too?
16:20 < fenn> the machine learning stuff?
16:21 < nmz787> probably
16:21 -!- preview [~quassel@203.167.243.242] has quit [Quit: No Ping reply in 180 seconds.]
16:22 < nmz787> how many years/lifetimes of study and theorization did it take for the first code compiler to exist?
16:23 < nmz787> that seems like it might be translated to how many it would take to learn how to fold ("compile") interesting proteins
16:23 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
16:23 < kanzure> folding isn't that big of a deal... just stick with the easy folds.
16:23 < kanzure> if all you have are sheets and coils... then use sheets and coils and stop complaining.
16:23 < kanzure> sheets and coils can probably do everything that globular proteins do, if you have the right functional amino acid sites.
16:24 < nmz787> sheets and coils are just one way to represent, usually because you aren't smart enough to think about the effects of electrons shifting around like an organic chemist or someone would
16:24 < nmz787> they don't really represent functions
16:24 < fenn> reproducible code is unfortunately not a priority for academia
16:24 < kanzure> there are a bunch of actual coil proteins
16:25 < fenn> i don't see any code (or input data) provided for the recent nature article "Planning chemical syntheses with deep neural networks and symbolic AI"
16:25 < fenn> compilers appeared within a decade or two
16:26 < kanzure> 06:45 < kanzure> "RetSynth: Solving all optimal retrosynthesis solutions using dynamically constrained integer linear programming" https://www.biorxiv.org/content/early/2017/11/21/223271
16:26 < kanzure> 06:45 < kanzure> https://www.github.com/sandialabs/RetSynth/
16:26 < fenn> i wish biorxiv would work with dillo
16:26 < kanzure> are you mad to ask for so much?
16:26 < fenn> always says '404'
16:26 < kanzure> you're crazy
16:27 < kanzure> this is an unreasonable request... they're going to just take biorxiv if we ask for that.
16:27 < kanzure> +away
16:27 < fenn> they are discriminating against people with lightweight browsers intentionally
16:27 < fenn> i think it's part of some half baked attempt to stop spiders
16:28 < kanzure> .g site:github.com retrosynthesis
16:28 < yoleaux> http://ipv6.google.com/sorry/index?continue=http://www.google.com/search%3F%26q%3Dsite:github.com%2520retrosynthesis%26btnI%3D&q=EhAqAQT4AgFyMAAAAAAACGrnGIn6xNYFIhkA8aeDS4ctDjmKjfKSQAizypMFCbDSmNH_MgFy
16:28 < kanzure> https://github.com/pandegroup/reaction_prediction_seq2seq
16:28 < kanzure> https://github.com/MarcStorm/Chemical-retrosynthesis
16:29 < kanzure> https://github.com/RamanLab/ReactionMiner
16:29 < kanzure> https://gist.github.com/kzfm/3402155
16:29 < fenn> do any of these programs actually work? as in, give you a synthesis pathway
16:29 < kanzure> https://github.com/mxk62/Retronet
16:30 < kanzure> yeah probably not :)
16:30 < fenn> this is a pretty tiny set of reactions https://github.com/MarcStorm/Chemical-retrosynthesis/blob/master/knowledgeBase.txt
16:30 < fenn> also it completely ignores stereochemistry and so on
16:31 < fenn> also chemical equations have equilibrium constants
16:32 < fenn> and enthalpies
16:34 < fenn> i know i'm being totally unreasonable again, asking for a retrosynthesis package that actually does something useful
16:34  * fenn crawls back into his hole
16:34 < kanzure> bottleneck is the database of organic chemistry reaction mechanisms. there's a few proprietary sources but nobody has scraped them yet.
16:34 < fenn> it's all scraped from the literature in the first place
16:35 < fenn> that was the point of the neural network paper
16:35 < fenn> .title https://www.nature.com/articles/nature25978
16:35 < yoleaux> Planning chemical syntheses with deep neural networks and symbolic AI | Nature
16:36 < kanzure> well i mean, retrosynthesis has been implemented for a few decades now, and there are companies that do this, which is why nobody has been working on the problem of collecting the data really....
16:36 < kanzure> there are entire databases owned by companies that just sit around typing up chemistry papers into a standard format for their database
16:36 < fenn> this sounds ripe for disruption
16:36 < kanzure> generally all the major chemistry journals are indexed and processed in this way
16:37 < fenn> i mean you can just mechanical turk it, and/or crowdsource it
16:37 < kanzure> "tell me... what good is a database, if you don't have access?"
16:37  * fenn chops off your ethernet cable
16:38 < kanzure> check yer inbox (sorry for the upside down email reading...)
16:40 < fenn> .title http://youtu.be/B203twyaMfM
16:40 < yoleaux> Starship Troopers (1997) Disable His Hand - YouTube
16:41 < kanzure> .title https://www.youtube.com/watch?v=ECe9Yqltw_w
16:41 < yoleaux> Species (1995) - Official Trailer (HD) - YouTube
16:42 < fenn> can you just dip pen polymerase onto the right location in a circuit?
16:42 < kanzure> i think it will incorporate multiple bases at a time
16:44 < fenn> polymerase
16:45 < fenn> like, take a RAM circuit or something, and put polymerase on the junctions between memory cells
16:45 < fenn> i'm sure there's a better IC for this
16:47 < fenn> don't need some super fancy $100k ASIC just to test out the idea
16:47 < nmz787> fenn: apparently dip-pen litho was all the rage at some trade shows 15-20 years ago... and little has been developed (or advertised about) since
16:48 < nmz787> fenn: from a guy in the FIB industry
16:48 < fenn> also you can get nmz787 to FIB you a demo circuit
16:48 < fenn> heh heh heh
16:48 < kanzure> hooking up polymerase into a circuit is not a problem
16:48 < nmz787> fenn: actually lately I've been thinking of decapping chips and mating them with some micro/nanofluidics
16:48 < nmz787> and then re-wirebonding to connect back up
16:48 < nmz787> or connect with microprober
16:49 < fenn> try putting some polymerase on them and see what sequences you get out
16:49 < nmz787> got a contact on a wire bonder refurbisher last night
16:49 < nmz787> actually I'd say hooking up the enzyme is a problem
16:49 < fenn> does decapping destroy gold wires? i thought it was just nitric acid
16:49 < nmz787> how do you do it repeatably?
16:49 < nmz787> i.e. the orientation
16:49 < fenn> you don't, it's random
16:50 < nmz787> azonenberg said QFNs are more and more using copper wires
16:50 < fenn> you could engineer some thing with positively and negatively charged domains linked to the polymerase
16:50 < nmz787> and they're wirebonded on the edges
16:50 < fenn> poo
16:51 < kanzure> hooking up polymerase ain't the problem.
16:51 < kanzure> https://github.com/kanzure/python-brlcad/pull/36
17:00 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has joined ##hplusroadmap
17:10 -!- hehelleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has joined ##hplusroadmap
17:14 -!- helleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has quit [Ping timeout: 260 seconds]
17:20 < l_wl> Do any of you live round Boston? I'm giving a talk at Harvard April 19th 10am y'all should drop by
17:21 -!- yashgaroth [~yashgarot@2606:6000:c308:f700:59ea:a32a:a020:b79c] has joined ##hplusroadmap
17:24 < kanzure> we will be in boston early may.
17:25 < l_wl> My talks  not that long
17:27 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 265 seconds]
17:27 -!- preview [~quassel@203.167.243.242] has joined ##hplusroadmap
17:40 -!- darsie [~username@84-114-73-160.cable.dynamic.surfer.at] has quit [Ping timeout: 260 seconds]
18:11 < streety> what are you talking about l_wl?
18:14 -!- aeiousomething [~aeiousome@124.123.14.99] has quit [Quit: leaving]
18:28 < l_wl> Wearable tech & off-ear hearables  aka bone-conduction & magneto restrictive transducers & all the cool products we're making at conduitsports.com
18:32 -!- preview [~quassel@203.167.243.242] has quit [Ping timeout: 264 seconds]
18:46 -!- Cory [~Cory@unaffiliated/cory] has quit [Ping timeout: 240 seconds]
18:47 -!- drewbot [~cinch@54.166.94.209] has quit [Remote host closed the connection]
18:54 -!- Cory [~Cory@unaffiliated/cory] has joined ##hplusroadmap
19:04 < kanzure> yashgaroth: for context, i think they are talking abuot biologically-relevant dna synthesis, but hteir proposal seems only relevant to dna data storage
19:04 < yashgaroth> ah okay
19:10 < kanzure> yashgaroth: i'd like to figure out a good experiment to suggest. but really nothing has occurred to me.
19:11 < kanzure> getting a really slow polymerase might be useful... if it's highly reliable.  and then you could wash out the dNTPs by flowing water over it... and send in new dNTPs for the next round whenever it incorporates a few seconds later..
19:11 < kanzure> and then you only need to make it promiscuous.. which is another challenge..
19:20 < yashgaroth> it's not gonna look much like a polymerase at that point
19:35 -!- emeraldgreen [~user@70.ip-145-239-90.eu] has joined ##hplusroadmap
19:41 -!- razzy [~user@unaffiliated/razzy] has quit [Read error: Connection reset by peer]
20:01 -!- justanotherus3r is now known as justanotheruser
20:14 < kanzure> hmph
20:34 < kanzure> if they want to do bluesky research... then let's look through the amino acids in phi29 and just pick some stuff.
20:34 < kanzure> might as well have a vaguely "rational" approach
20:35 < yashgaroth> quite a few mutants on uniprot http://www.uniprot.org/uniprot/P03680
20:38 -!- jtimon [~quassel@142.29.134.37.dynamic.jazztel.es] has quit [Ping timeout: 276 seconds]
20:40 < kanzure> with a phi29 pol that can be biased to incorporate a single nucleotide.. you only need two of those, really.. you do RCAs to produce DNA (with restriction sites...) based on the template + the one change.. and then you re-circularize the generated DNA and run it through the other modified phi29 pol RCA reaction to get the other data.... but alignment is a problem.. and timing.. unless you do ...
20:40 < kanzure> ...sequencing-by-synthesis fluorescence.... yeah this is too much shit already.
20:42 < kanzure> with this "two step" system, you can constantly go back and forth between the different phi29 pol RCA reactions until you get the exact sequence you want
20:42 < kanzure> (during each RCA reaction you would be doing some interference to force the polymerase to incorporate the other nucleotide type, the one that it's not seeing on the template, for positions where you want to write that nucleotide)
20:45 < yashgaroth> I'm stealthily trying to direct it back to DNA data storage, as I doubt phi29 will be the ideal for synthesis...probably best to start with some super well-characterized generic pol for that
20:45 < kanzure> btw i haven't had them working on dna data storage
20:46 < kanzure> i think our technique is relatively good and right now the bottleneck is machine scale-up
20:47 < kanzure> if they want to work on dna data storage then i guess i could consider that
20:47 < yashgaroth> is that with dpo4 or phi29?
20:47 < kanzure> phi29 pol so far
20:48 < yashgaroth> wild type, no exonuclease mutant?
20:48 < kanzure> wild type
20:48 < yashgaroth> dang, well I imagine it'd be even better with one of those mutants
20:52 < kanzure> one of the other tricks up my sleeve is that academia often focuses on comprehensibility... e.g. trying to isolate the mechanism by which things work.
20:52 < kanzure> but if you're willing to be sloppy... you can do things like, uh, in vivo RCA and have like 50 different mutant pol genes in there.
20:52 < kanzure> and you only care about one of the polymerases getting mutated and being selected... but you also don't care if there's a bunch of other noise going on.
20:53 < kanzure> "who knows why it works, it just does" situations
20:53 < yashgaroth> if it works but it's sloppy, it still works
21:03 < kanzure> feel free to send them the two-RCA-step thing i mentioned above.. i'm going to bed. night.
21:03  * night 
21:07 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has quit [Remote host closed the connection]
21:07 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has joined ##hplusroadmap
21:12 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has quit [Ping timeout: 268 seconds]
21:28 -!- yashgaroth [~yashgarot@2606:6000:c308:f700:59ea:a32a:a020:b79c] has quit [Quit: Leaving]
21:29 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has quit [Ping timeout: 264 seconds]
21:53 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has joined ##hplusroadmap
22:03 -!- wrldpc1 [~ben@softbank126225109178.bbtec.net] has quit [Ping timeout: 276 seconds]
22:06 -!- MrHindsight [~2020@185.174.139.20] has joined ##hplusroadmap
22:06 -!- MrHindsight [~2020@185.174.139.20] has quit [Changing host]
22:06 -!- MrHindsight [~2020@unaffiliated/capthindsight] has joined ##hplusroadmap
22:08 < MrHindsight> disrupt this  https://arstechnica.com/tech-policy/2018/04/curing-disease-not-a-sustainable-business-model-goldman-sachs-analysts-say/
22:09 -!- CaptHindsight [~2020@unaffiliated/capthindsight] has quit [Ping timeout: 240 seconds]
22:43 -!- emeraldgreen [~user@70.ip-145-239-90.eu] has quit [Quit: Leaving.]
22:43 -!- emeraldgreen [~user@70.ip-145-239-90.eu] has joined ##hplusroadmap
22:50 < fenn> i still think polymerase ratcheting is important; doesn't it proceed as a poisson process? how do you know when to kick it when it randomly incorporates a base
22:51 < fenn> it might be as easy as causing mutations until the enzyme stops incorporating anything under normal conditions, and then zapping it with a voltage
22:51 < fenn> i don't see any reason why zapping an enzyme with different voltages would cause it to incorporate different bases, but i could be wrong
22:51 < fenn> more likely the direction (vector) of the electric field is important
22:52 < fenn> but that particular experiment requires fancy custom nano-circuitry
22:53 < l_wl> Wait do left handed proteins respond differently than right-handed ones to electricity
22:54 < fenn> or at least custom normal-circuitry, i.e. a reticule (+ sign with a hole cut out the middle) with the polymerases in the center
22:54 < fenn> l_wl: there are no left handed proteins
22:55 < l_wl> Or compounds
22:55 < fenn> stereoisomers are detected by the direction they rotate polarized light
22:56 < fenn> D-glucose rotates clockwise, L-glucose rotates counterclockwise
22:58 < l_wl> that makes sense cool. Why are there no left handed proteins
22:58 < fenn> because life only make a particular stereoisomer of each amino acid
22:58 < fenn> except a few which are too simple to have stereo centers, like glycine
22:59 < l_wl> Is this because dna has a certain direction?
22:59 < fenn> i don't know
23:00 < fenn> the RNA hypothesis for the origin of life says that early enzymes were made of RNA, which is handed
23:00 < fenn> the first amino acids would have been made by those enzymes
23:01 < l_wl> Ah so then if we all base our rna from that the other proteins would follow
23:06 < fenn> apparently it's an unsolved question
23:08 < fenn> i think it's just network effects, like why everyone uses the same alphabet
23:09 < fenn> then you throw out 99% of the population and duplicate the remainder until you're back to where you were, repeat that a few million times and you get everyone using exactly the same spelling
23:10 < fenn> a cell that uses a different alphabet can't get new genetic material from other cells
23:10 < l_wl> cool - makes sense. So why are you electrocuting bio-cells
23:11 < fenn> to encode data into the DNA they are producing
23:11 < fenn> ideally we'd be able to make the exact DNA sequence we want
23:12 < fenn> but that's hard
23:12 < l_wl> Data just to store info? Or to express certain things like what cripsr is for
23:12 < fenn> both
23:12 -!- ExeciN [ExeciN@bnc.stormbit.net] has quit [Ping timeout: 245 seconds]
23:13 < fenn> perhaps blasting the enzymes with right- or left- circularly polarized light would cause them to incorporate different bases
23:14 < l_wl> I don't see the reason for the first. But for the latter how are you going to insert DNA elements without providing anything (apart from electricity)
23:14 < l_wl> oh
23:14 < fenn> it's copying a template strand like AAAAAAAA
23:15 < fenn> storing computer data on DNA is useful because it's really really compact
23:15 < fenn> also you can pull out sequences of interest with PCR
23:15 < fenn> it lasts millions of years in easily achieve storage conditions
23:16 < fenn> the copying process is very low energy and extremely inexpensive
23:17 < l_wl> But very slow. Also it lasts millions of years how - are you assuming the organisms are replicating?
23:18 < fenn> it's not slow. a single polymerase can copy DNA at 1000 bases per second, and you can have millions per nanoliter
23:18 < fenn> it lasts millions of years by sitting there and not degrading
23:18 < fenn> viable bacteria have been recovered from inside salt minerals
23:19 < fenn> it may require some protein to achieve that level of reliability
23:19 < l_wl> Hmm anyway manipulating an enzyme with electricty seems cool and useful
23:19 < fenn> you don't even need to repair it because we have error correction codes
23:23 < l_wl> Samumed (https://www.samumed.com/default.aspx), which is a company that has invented 8 successful cancer drugs, as well as a trial drug for degenerative disc disease. The way it works: they pump extremely high doses of chemotherapy medicine into your blood stream, at a level that would normally kill you. But its contained in nanoshells. When it gets to the area of the cancer, a laser tool is used to crack the shel
23:23 < l_wl> ls open and deliver very localised cancer treatment. The really cool thing is that instead of using some specialised laser tool which they could sell for a fortune, they designed it to work with any ultrasound wand.
23:23 < l_wl> https://l.facebook.com/l.php?u=https%3A%2F%2Fwww.samumed.com%2Fdefault.aspx&h=ATPrhqLBoBcmRy7MG86Qv9KDfHgjyKQ8Tmwdp8fJVNRX6QKXua9cPnS-CievIXXA350E1iPp1RaJCIxrtWJv_f2IfQ8sFz-NfXDnGWEaZ-hCIP1Bmn4AsdK3
23:24 < fenn> are you sure it's a laser
23:24 < l_wl> No my friend wrote that.
23:25 < fenn> this website is too much of a pain in the ass for me to want to care
23:25 < l_wl> will check 4 you
23:25 < fenn> also there are a million therapies for cancer
23:25 < fenn> the most promising ones harness the immune system to attack the cancer
23:26 < fenn> that way it doesn't come back
23:26 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has quit [Quit: It's good to know that theirs will suffer EVEN MORE than ours will.]
23:29 < MrHindsight> there are better, faster and smaller molecular storage mechanisms, DNA storage is just a fad
23:30 < MrHindsight> and now easily rewritten/edited
--- Log closed Sat Apr 14 00:00:47 2018