--- Log opened Fri Mar 01 00:00:13 2024
03:06 < hprmbridge> kanzure> "Review of programmable RNA-targeting systems as well as modification enzyme-based effector proteins" https://www.nature.com/articles/s41589-023-01531-y
03:08 < hprmbridge> kanzure> "Phage-assisted evolution of highly active cytosine base editors with enhanced selectivity and minimal sequence context preference" https://www.nature.com/articles/s41467-024-45969-7
03:22 -!- Croran [~Croran@user/Croran] has quit [Ping timeout: 256 seconds]
03:24 -!- Croran [~Croran@user/Croran] has joined #hplusroadmap
03:26 -!- ike8 [e8f913dbdf@irc.cheogram.com] has quit [Ping timeout: 252 seconds]
05:17 < hprmbridge> kanzure> https://github.com/nanographs/Open-Beam-Interface
05:38 < kanzure> germline genetic engineering act (not really, it's just about IVF in general, but it does not seem to impose limitations on germline) (not passed) https://www.congress.gov/bill/118th-congress/house-bill/7056/text?s=1&r=7
05:42 < kanzure> hmm "GOP senator blocks IVF protection bill because it could lead to human-animal hybrids"
05:43 -!- ike8 [e8f913dbdf@irc.cheogram.com] has joined #hplusroadmap
05:43 < hprmbridge> soul_syrup> @jrayhawk arent most of the anit eugenicists just anti science also?
05:44 < hprmbridge> soul_syrup> theyre just wokescience propagandists with a strong bitter note
05:46 < jrayhawk> Nope, the negative affect on eugenics is something trained into everyone in most western countries independent of political valence
05:47 < hprmbridge> soul_syrup> "independent of political valence" "trained into everyone in most western countries"
05:47 < hprmbridge> soul_syrup> this
05:48 < jrayhawk> I mean of traditional hierarchicalist/antihierarchicalist valence that "wokescience" is invoking
05:48 < hprmbridge> soul_syrup> still
05:49 < hprmbridge> soul_syrup> western education = wokescience pretty much
05:49 < hprmbridge> soul_syrup> media/education/law/etc
05:49 < jrayhawk> There is certainly an additional "bioethicist" faction worth thiking about and an additional antihierarchicalist faction worth thinking about when building rhetoric
05:49 < hprmbridge> soul_syrup> if you cant burn it, leave it
05:50 < hprmbridge> soul_syrup> thats how a lot of people think about broken systems
05:50 < hprmbridge> soul_syrup> science is alive and thriving, just not so much over there
05:52 < jrayhawk> you have to engage with your rhetorical enemies in order to avoid getting consumed by confirmation bias
05:52 < hprmbridge> soul_syrup> dont you think that is a little naive?
05:52 < hprmbridge> soul_syrup> I totally understand it, but maybe its a waste of time to talk to retards?
05:56 < jrayhawk> certainly there is a cost-benefit curve, but every lasting social institution relies on adverserial collaboration for dynamicity. disaster occurs when a political faction is allowed to "win" to the point where opposing viewpoints can no longer be represented.
05:56 < hprmbridge> soul_syrup> constructive debate is  not the same
05:56 < hprmbridge> soul_syrup> wokescience people just waste your time and energy
05:58 < hprmbridge> soul_syrup> the point of the constructive debate is not to destroy a society/nation, whereas wokescience aims to do exactly that
05:58 < jrayhawk> Critical theory is not useless. Sometimes hierarchies stagnate and corrupt. Sometimes important social forces really do come down to will-to-power.
05:58 < jrayhawk> The problem with critical theory is when it becomes reductive in claiming *everything* is about those things.
05:59 < hprmbridge> soul_syrup> critical theory is a degeneracy of critical thinking
06:01 < hprmbridge> soul_syrup> the idea of racially homogeneous city states is a product of critical thinking
06:01 < jrayhawk> Their instinct is to attack hierarchies regardless of whether or not it creates good outcomes. Some people similarly have an instinct to support hierarchies regardless of whether or not it creates good outcomes. Whether or not these people are virtuous or unvirtuous depends upon the context they're placed in.
06:01 < hprmbridge> soul_syrup> the idea of multicultural race war is the product of critical theory
06:02 < hprmbridge> soul_syrup> their instinct is designed
06:02 < jrayhawk> I am sorry, you've lost me, there. Multi-faction race wars are as old as human history.
06:02 < hprmbridge> soul_syrup> social weapons
06:02 -!- adlai [~adlai@user/adlai] has quit [Quit: WeeChat 4.1.1]
06:02 < hprmbridge> soul_syrup> multi faction race wars are part of humankind
06:02 < hprmbridge> soul_syrup> but why create it on purpose inside cities
06:03 < hprmbridge> soul_syrup> segregation is the source of diversity
06:04 < jrayhawk> There's, again, a tradeoff between social trust plus stagnation and dynamicity plus strife in multiculturalism.
06:05 < hprmbridge> soul_syrup> mix all colors and what do you get
06:05 < hprmbridge> soul_syrup> there needs to be some form of homogeneity
06:05 < jrayhawk> decreased trust, greater dynamicity. Whether that's a good thing or not depends upon context and the value system involved.
06:06 < hprmbridge> soul_syrup> complete self annihilation is a terrible value system
06:06 < hprmbridge> soul_syrup> systems, balance, self segregation
06:06 < hprmbridge> soul_syrup> harmony
06:07 < kanzure> i want to torpedo this. can we go back to work.
06:07 < hprmbridge> soul_syrup> network dynamics
06:07 < jrayhawk> that's fine
06:07 < hprmbridge> soul_syrup> 👍
06:48 < L29Ah> why are americans so obsessed about races?
06:51 < TMA> because they have much more recent history of slavery where the division of free/slave was largely coincident with race
06:53 < kanzure> yawn.. can we get back to work please.
06:54 < hprmbridge> soul_syrup> probably because there is no american race and they feel sort of non ethnic
06:54 < TMA> nope. it's Friday already. it is really hard to concentrate on _work_ when the weekend is near
07:01 < kanzure> work harder
07:04 < L29Ah> how do i work harder on figuring out what to work on?
07:09 < kanzure> you have to run twice as fast just to keep up with as fast as once as fast
08:02 < fenn> the RIMS1 screenshot was from https://www.astralcodexten.com/p/links-for-february-2024 and also links to https://www.astralcodexten.com/p/evolution-explains-polygenic-structure and the same ncbi link above
08:10 < fenn> "the pinnacle of air-breathing engine design, the SR-71 Blackbird." (designed in 1956) and if you believe that i have an airport to sell you
08:11 < fenn> 1958*
08:36 < hprmbridge> soul_syrup> @fennfoot it makes sense genetically to not get rid of it since it does improve chance for a group to invent new things
08:36 < hprmbridge> soul_syrup> i mean the closeness of mad scientist and skizo ...
08:43 < fenn> i think it's totally plausible that schizophrenia is viral infection sequelae
08:43 < fenn> there are many sub-clinical viruses that go mostly unnoticed, unprevented, and unmonitored
08:46 < fenn> "half of people are at the 50th percentile of schizophrenia risk and nothing bad happens to them." *until they get schizophrenia*
08:47 < fenn> is there a more emphatic facepalm emote i can use
08:47 < fenn> scott you ought to know better!
08:49 < L29Ah> nice filler, i wonder if a LLM can be tuned to insert such smartass statements into articles so they are as long as Scott's
08:50 < jrayhawk> "the genes load the gun, the environment pulls the trigger" would've been a lot shorter
08:58 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
08:59 -!- cthlolo [~lorogue@77.33.24.3] has joined #hplusroadmap
09:15 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
09:18 < fenn> i feel the need to defend what i said
09:19 < fenn> scott is advocating the idea of choosing embryos with 50th percentile schizophrenia risk, on the unfounded theory that schizophrenia genes *might* provide some as-yet unknown protective effect
09:19 < fenn> this would be committing a generation of people to a disease burden that could have been avoided
09:21 < fenn> chesterton's fence doesn't apply to evolution. we've changed so many variables in the environment already, with agriculture, the industrial revolution, global travel, chemistry
09:21 < fenn> if humans are choosing the genes, it would be "humans load the gun, the human-constructed environment pulls the trigger"
09:24 < fenn> it's like saying "it can't be bad to buy IBM"
09:24 < fenn> well actually, it can
09:26 < fenn> if you load the gun and you pull the trigger, you're culpable
09:48 < TMA> 'you have to run twice as fast just to keep up' is exactly the good reason for wanting to die sooner rather than later
09:53 < fenn> that's the depression talking
09:53 < fenn> you can do twice as much now, because it's the future
09:54 < TMA> it is the depression. I am already tired of constantly doubling my running speed
09:55 < fenn> go live on one of those organic human farms
09:56 < fenn> i suspect you will be just as tired while rooted in the ground
09:57 < fenn> we can give you green skin and you can be a plant
10:00 < L29Ah> TMA: TMA is probably a passable antidepressant btw
10:00 < TMA> not-being sounds still as the most appealing options of those mentioned so far; I am an atheist, there is no afterlife for me, for which I am grateful to all the nonexistent gods and Gods [irrespective of their gender, number]
10:01 < TMA> L29Ah: :-)
10:02 < L29Ah> TMA: you can also pick dissociative fugue or dissociative identity disorder
10:02  * TMA has been told that controlled amounts of interaction with me has positive impact on certain class of humans
10:03 < L29Ah> https://psychonautwiki.org/wiki/TMA very
10:03 < TMA> ;-P
10:36 < hprmbridge> Eli> We evolved to believe in a creator. It's unique to humans.
11:34 -!- cthlolo [~lorogue@77.33.24.3] has quit [Quit: Leaving]
11:51 < hprmbridge> kanzure> RIMS1 authors now recant: https://twitter.com/richardfuisz/status/1763366885188391082
11:52 < superkuh> Aw.
12:57 < fenn> https://www.courthousenews.com/elon-musk-sues-openai-over-ai-threat/
12:58 < fenn> RIMS1 doesn't make you blind? great! what about the 20 IQ points?
12:59 < fenn> "The paper only addressed the blind part, not the wizard part…. So you can still keep the dream alive"
13:02 < fenn> this doesn't seem possible: "99% of all single-nucleotide substitutions compatible with life have been present in at least one living human for about 200 years" https://nitter.privacydev.net/pic/orig/media%2FGHmI9i3WEAAXy_9.jpg
13:05 < fenn> are the religious people against breeding humans like cattle? it seems to me that was the standard practice for much of human history
13:05 < fenn> if there's a family with a special +20 IQ points gene we need to be breeding the shit out of them
13:05 < Betawolf> that doesn't sound right. We were a lot more selective with cattle breeding.
13:06 < fenn> aristocracy did (and does) all sorts of horse trading
13:06 < fenn> i'm not talking about the commoners
13:07 < fenn> i'm not saying it worked, mind you, but it is a tradition
13:08 < Betawolf> The problem is that the weak bloodlines still continued, and there was a lot more random loss of stock
13:10 < Betawolf> with a few exceptions, we didn't have one impressive specimen produce 30+ children and keep only the impressive ones in the gene pool, it was more like 5-10 and a good number of them would reproduce regardless of quality
13:13 < fenn> surely they saved the good princesses for the good princes?
13:14 < fenn> it is possible to lose one's title through mismanagement or other deeds
13:14 < hprmbridge> soul_syrup> aristocracy look like literal inbreeds though
13:15 < fenn> i mean, they are, so that checks out right?
13:15 < hprmbridge> soul_syrup> exactly lol
13:15 < hprmbridge> soul_syrup> well technically in japan they had selective breeding through pressured suicide
13:16 < fenn> how many extra IQ points does it take before you'll accept blue blood as the price to pay
13:16 < hprmbridge> soul_syrup> and in germany there was a law that made race mixing punishable by death
13:16 < hprmbridge> soul_syrup> -50 iq
13:16 < fenn> huh?
13:17 < fenn> you want blue blood and are willing to pay 50 IQ for it?
13:17 < hprmbridge> soul_syrup> yes i want to be dumb and ugly
13:17 < fenn> well that's very achievable
13:17 < hprmbridge> soul_syrup> "i just want to fit in "
13:19 < hprmbridge> soul_syrup> random question but does anyone here want to conduct remote experiments that dont need too expensive equipment, but are illegal in their country?
13:20 < hprmbridge> soul_syrup> I'll happily help, just for scientific progress
13:20 < fenn> i'd request that this sort of negotiation happen in PM
13:20 < hprmbridge> soul_syrup> ah ok anyone intereste pm me
13:20 < hprmbridge> soul_syrup> or on telegram if you feel better
13:21 < fenn> thanks for offering
13:22 < hprmbridge> soul_syrup> my telegram if you wanna pm Apfjsibgjwjsjjx728gsj73jej284sn
13:24 < fenn> material transfer agreements are a real choke point
13:25 < hprmbridge> soul_syrup> specific?
13:25 < fenn> we've built up this whole edifice of science the relies on access to proprietary cell lines like ATCC
13:25 < fenn> you have to sign this document that says you won't send them to anyone who hasn't signed the document
13:25 < hprmbridge> soul_syrup> so?
13:26 < fenn> well, it makes it hard to reproduce science outside of "the system"
13:26 < hprmbridge> soul_syrup> where is the regulatory inspection?
13:26 < hprmbridge> soul_syrup> well tell me what stuff you want access to and I'll apply for it
13:26 < hprmbridge> soul_syrup> lol
13:27 < fenn> e. coli
13:27 < hprmbridge> soul_syrup> thats super easy
13:27 < fenn> :\
13:27 < hprmbridge> soul_syrup> if you want tuberculosis i can get a sample for you lol
13:27 < hprmbridge> soul_syrup> incase youre interested in that type of bio stuff
13:27 < fenn> i don't want tuberculosis
13:28 < fenn> nobody should want tuberculosis
13:28 < hprmbridge> soul_syrup> you can do a lot of fun stuff with it lol
13:28 < fenn> please explain
13:28 < hprmbridge> soul_syrup> that i will not <:Nervous:1192926522546323606>
13:29 < fenn> maybe you've got the wrong idea. we're trying to make people live longer here
13:31 < fenn> [REDACTED]
13:32 < fenn> could be useful for phage therapy
13:32 < fenn> since apparently NOBODY IS FIXING IT
13:32 < fenn> ffs if there were actually some adults in charge this totalitarian over-reach might make sense
13:35 < fenn> there are more and less dangerous strains, i.e. with and without toxin genes, antibiotic resistance genes. phages could be developed with knock-out strains to maintain diversity, but i think it's still a dangerous and foolish thing to mess around with
13:44 < L29Ah> 22:19:40]<hprmbridge> soul_syrup> random question but does anyone here want to conduct remote experiments that dont need too expensive equipment, but are illegal in their country?
13:44 < L29Ah> yes very, no discord tho
13:51 < geneh2_> reminder it is illegal to distribute code for integrating certain equations with spherical symmetry in the USA
13:52 < fenn> guess i'll shred all my general purpose computers
14:00 < L29Ah> remember it is illegal to enter http://superkuh.com/ if you're under 90 years old
14:19 -!- darsie [~darsie@84-112-12-36.cable.dynamic.surfer.at] has quit [Ping timeout: 246 seconds]
14:35 < hprmbridge> richardfuisz> The basic conservative math intuition here:
14:35 < hprmbridge> richardfuisz>
14:35 < hprmbridge> richardfuisz> 8B people
14:35 < hprmbridge> richardfuisz> 50 de novos SNPs per generation
14:35 < hprmbridge> richardfuisz> 9B possible SNP mutations
14:35 < hprmbridge> richardfuisz> 8B*50/9B =50 people currently alive today with each SNP as a de novo mutation.
14:37 < hprmbridge> richardfuisz> I don’t have UKBB access, but after posting I heard today from someone who does and they couldn’t get rims1 to replicate. So I declare it tentatively dead
14:39 < geneh2_> born secret laws make it really difficult to legally analyze some advanced propulsion concepts like the nuclear salt water rocket
14:43 < fenn> oh hello
14:47 < fenn> richardfuisz: i don't pretend to know statistics, but my pet robot pretends to: https://pastebin.com/raw/5knsKbLs
14:49 < fenn> this is totally a homework problem
14:53 < fenn> post hot take on twitter, get the internet to correct your mistakes, submit the best argued answer
14:54 < fenn> it's not academic dishonesty if you cite your sources
15:10 -!- justanot1 [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
15:10 < fenn> richardfuisz the basic error in your explanation is that there isn't a queue handing out unique SNPs until they're all gone, they will overlap. it's like the birthday problem, and you're trying to find a room of people with every birthday. it has to be much more than 365 people. i fiddled with the numbers in cochran's sample size formula (not that i understand it) and we can say with ~50%
15:10 < fenn> confidence that 95% of the SNPs exist, assuming they're uniformly distributed
15:11 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 260 seconds]
15:18 < fenn> (0.67^2 * 0.95 * 0.01) / (1-.95)^2 * 9billion*.95 = 14 billion
15:19 < fenn> i guess i should be considering the cumulative population, which is like 100 billion
15:23 < hprmbridge> richardfuisz> For most purposes it’s materially the same is 95% vs 100% SNP coverage
15:24 < hprmbridge> richardfuisz> Yeah. Of course then you do have some drift / dropout
15:25 < fenn> with 200 billion births, we still only have 50% confidence that there's 98% of all possible SNPs
15:26 < fenn> sorry i should have just said:
15:26 < fenn> with 126 billion births, we still only have 50% confidence that there's 97% of all possible SNPs
15:27 < fenn> and i missed the 0.01 above, it should be 0.05 which raises the number on the right side of the equation by 5x
15:29 < hprmbridge> richardfuisz> I think it’s also fair, in this context, to ask “have we saturated the mutations that are compatible with life”
15:29 < hprmbridge> richardfuisz> Which is what the original graph does.
15:30 < hprmbridge> richardfuisz> I guess, to put this a different way — what is a context where you’d make a different decision if there was 50% confidence of 97% coverage vs 99% coverage vs 90% coverage?
15:31 < fenn> well the whole thing is silly anyway because evolution happens
15:31 < fenn> it's extremely not evenly distributed
15:32 < fenn> this is the paper the graph is from https://www.cell.com/trends/molecular-medicine/abstract/S1471-4914(16)30122-8
15:32 < hprmbridge> richardfuisz> I mean the underlying point here that I actually find interesting, is that most VUS should have a population out there somewhere that we could sequence & Eval
15:32 < hprmbridge> richardfuisz> And if a baby is born with a de novo mutation, the odds are very low that that mutation has never happened before. That gives some indication for how to approach treatment in those cases.
15:33 < fenn> i suppose it would be interesting in the same way looking at aircraft returning with bullet holes in them tells you what's important
15:33 < fenn> we keep sending the aircraft out without patching up the bullet holes
15:34 < hprmbridge> richardfuisz> Yeah — if you maintain / design aircraft for a living, that seems pretty important info
15:36 -!- justanot1 [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Remote host closed the connection]
15:36 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
15:43 < fenn> i'm not sure if there is any difference between "conserved sequences" and "vital sequences"
15:45 < fenn> i could imagine special protection mechanisms that hang around strongly conserved regions and repair things as soon as possible, conserving those sequences, and meanwhile other just-as-vital sequences get mutated and the organism dies
15:46 < fenn> sort of like putting armor on the airplane with the bullet holes, maybe the armor is in the wrong place
15:46 < fenn> if there's no armor (special DNA repair), then the only way the conserved sequences are conserved is because any time they changed the organism died or lost lots of reproductive fitness
15:55 < fenn> oh you divide by 50-100 at the end, whoops
15:55 < fenn> it's still not quite there but at least it's not a totally crazy mismatch between the graph and what i calculated
15:56 < fenn> i still think the graph itself is wrong
15:58 < fenn> not a totally crazy mismatch between the "over 99%" claim and what i calculated
16:00 < fenn> the graph implies we hit 99% somewhere around 1800
16:00 < hprmbridge> richardfuisz> i think that year 1800 figure is mostly driven by cumulative mutation
16:01 < hprmbridge> richardfuisz> integrating the pop size pre 1800, with 50 de novos, gets you somewhere
16:02 < hprmbridge> richardfuisz> now the one caveat that no one has mentioned on the tweet thread or anywhere i've seen, is that de novo mutation rate actually varies a fair bit -- young parents have lower de novos than old parents for example. that 50 de novos per year may not have held in the year 1200
16:02 < fenn> yes, it takes ~45 billion births which would be about half the area under the curve which looks to be around 1800
16:03 < fenn> i'm still not sold on the idea that suboptimal but not immediately deadly SNPs will stick around for hundreds of years
16:04 < fenn> at least i proved the simplifed statistical model to myself
16:04 < fenn> sorry for the noise
16:05 < fenn> half the mutations occurred in the last 200 years
16:07 < hprmbridge> richardfuisz> i mean, i think most GWAS show that mildly suboptimal SNPs not only survive, but actually have pretty high MAF
16:38 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Ping timeout: 255 seconds]
17:32 -!- ike8 [e8f913dbdf@irc.cheogram.com] has quit [Remote host closed the connection]
18:00 < fenn> i keep watching these demos over and over and i really think this is a breakthrough. we have a minimum viable AI virtual avatar stack now, with bark, whisper, and https://humanaigc.github.io/emote-portrait-alive/
18:01 < fenn> it needs some sort of emotional valence understanding system still. i'm not even sure how to formulate that in words
18:02 < fenn> when you say something to whisper it gets turned into text. there's only so much emotional side channel info you can encode in the speechML markup that we have so far
18:03 < fenn> and whisper isn't even trained to understand emotion (yet)
20:24 -!- mxz [~mxz@user/mxz] has quit [Ping timeout: 240 seconds]
22:30 -!- mxz [~mxz@user/mxz] has joined #hplusroadmap
23:40 < hprmbridge> Eli> Apparently, astaxanthin is an OTC supplement that increase lifespan 10% in mice. This is according to the ITP, so it's likely a real signal. We could come up with a lot of hypothesis for why this is, but we don't really know at this point. Anyway, we do know it's pretty good for humans already. And if there's a potential lifespan increase, then why not take it?
23:46 -!- darsie [~darsie@84-112-12-36.cable.dynamic.surfer.at] has joined #hplusroadmap
--- Log closed Sat Mar 02 00:00:15 2024