--- Log opened Sat Sep 13 00:00:38 2025
00:01 < fenn> sounds like more work than inserting a stable landing pad region
00:02 < fenn> would you rather have the option to be "cured" of the HAC, or the option to transmit upgrades to offspring the old fashioned way
00:03 < fenn> maybe once the upgrades have been proven it can be baked in
00:08 < fenn> unfortunately i don't see how HACs solve the gene therapy delivery problem
00:09 < fenn> cells with version 1 express a surface receptor version 1. the vector delivering version 2 binds to receptor version 1, and includes code to inactivate receptor version 1 and express receptor version 2
00:10 < fenn> you could do this without HACs of course. i guess the benefit would be that HAC version 2 would displace HAC version 1 without any chance of off-target effects?
00:11 < fenn> also this "hardware unlock" should be induced via a very noticeable intervention performed by the human being upgraded, so it can't be done maliciously without their knowledge (and ideally without their consent, but i don't know how to make that work)
00:12 < fenn> if you literally break out in green spots you know your upgrade mechanism has been activated
00:13 < fenn> and take the antidote
00:13 < fenn> we shold strive for better security than shitty IoT devices
01:11 < hprmbridge> .monokhrome> has anyone tried using killer t cells for gene therapy, have them inject a  gene therapy payload to fix the faulty gene when they detect the faulty protein on the cell surface instead of granzymes?
02:07 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
02:07 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
02:53 < hprmbridge> kanzure> yes see https://diyhpl.us/wiki/cell_therapy
02:58 < hprmbridge> kanzure> you might be interested in the neochromosome proposals either in the logs or idk if it's on the wiki here https://diyhpl.us/wiki/genetic-modifications btw embryo microinjection is inappropriate for HACs, just so it in a stem cell culture, do sequencing and verification because we are not morons, and inject modified cell nucleus into enucleated egg or embryo.
02:59 < hprmbridge> kanzure> hmm it's not on that page. you are all fired!
03:00 < fenn> what's the t-cell vector called? i don't see anything particularly relevant on the cell_therapy wiki page
03:01 < hprmbridge> kanzure> ugh really
03:02 < hprmbridge> kanzure> church has a similar scheme, again in ancient logs here, for killer T cell destruction of unmodified spermatagonians to appease the catholics
03:19 < hprmbridge> kanzure> here is a review of different (mainly CRISPR?) gene therapies for the different T cell therapies https://cmbl.biomedcentral.com/articles/10.1186/s11658-024-00561-1 including "CD8+ cytotoxic T cells and CD4+ helper T cells, primarily including Th1, Th2, Th17, and Treg, as well as a small subset known as γδ T cells characterized by the presence of γ and δ chains in their T cell receptors (TCR)."
03:28 < hprmbridge> kanzure> there has also been work on synthetic receptors for T cell engineering https://www.nature.com/articles/s41423-025-01304-8 and things like synNotch receptors for programmable antigen targeting, or multi-receptor logic gating see section "synthetic immune-system engineering" on last wiki page link, ADR, B cell receptor engineering
03:29 < hprmbridge> kanzure> also section "Cell–cell sensing & programmable logic" https://diyhpl.us/wiki/morphogenesis/
03:29 < hprmbridge> kanzure> maybe we need a synNotch page too
03:30 < fenn> i think maybe you didn't understand the question? the goal is to use a t-cell as a gene therapy vector
03:31 < hprmbridge> kanzure> no he said killer T cell, that's a specific type
03:33 < hprmbridge> kanzure> ok fine.
03:37 < hprmbridge> kanzure> what i am not clear about is whether to link the trogocytosis article https://www.biorxiv.org/content/10.1101/2025.03.12.642522v1.full or not, because other cells do it via microtube transfer, not lipid membrane exchange, right?
03:52 < hprmbridge> kanzure> "In vivo human T cell engineering with enveloped delivery vehicles" https://www.nature.com/articles/s41587-023-02085-z  (particles cloaked in cellular membrane fragments including antibody-derived single-chain variable fragments (scFvs) for cell targeting and VSVGmut for endosomal fusion or cell fusion)
03:55 < hprmbridge> kanzure> "Syncytin-mediated open-ended membrane tubular connections facilitate the intercellular transfer of cargos including Cas9 protein" https://elifesciences.org/articles/84391 via tunneling nanotubes
04:05 < hprmbridge> kanzure> "Beyond synapses: cytoplasmic connections in brain function and evolution" https://onlinelibrary.wiley.com/doi/full/10.1111/brv.70034 including "direct cellular communication through modalities such as intercellular bridges (IBs) formed during incomplete cytokinesis, de novo tunnelling nanotubes (TNTs), and cytoplasmic connections arising from cell–cell fusion."
04:13 < hprmbridge> kanzure> "cell penetrating peptides (CPPs), CPP additives, bacterial toxins, and contractile injection systems"
04:16 < hprmbridge> kanzure> here's one that does translocation of Cas9 using "reengineered anthrax toxin" https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2021.770283/full
04:33 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1416386184472694894/image0.jpg?ex=68c6a7ed&is=68c5566d&hm=f16a39ba4763a87f7e49d22c6ae6b5c82aee5aa597b52bcd9c3ebfa598db5133&
05:56 < hprmbridge> kanzure> "Targeted cell-to-cell delivery of protein payloads via the granzyme-perforin pathway" https://pmc.ncbi.nlm.nih.gov/articles/PMC5700818/ (2017)
06:02 < hprmbridge> kanzure> " size of these pores is generally larger than 10 nm"
06:32 < hprmbridge> kanzure> "interoception / immunoception and the continuous bidirectional flow of information between the brain and the immune system... The concept of an immunengram suggests that the brain can form a specific neuronal trace in response to immunological events, which can be retrieved upon reactivation of the same neuronal ensembles. However, in contrast to the neuronal engram, which is specific to the
06:32 < hprmbridge> kanzure> brain, and for which neuronal activity is sufficient to manifest the required behavior, we suggest that in the immunengram, the trace is not limited to the neuronal component. It involves changes in tissue cells in the periphery and, potentially, in specific immune clones." mainly via the insular cortex.
06:33 < hprmbridge> kanzure> "Insular cortex neurons encode and retrieve specific immune responses" https://www.cell.com/cell/fulltext/S0092-86742101223-X
06:44 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
06:45 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
08:03 < hprmbridge> kanzure> "Targeted delivery of diverse biomolecules with engineered bacterial nanosyringes" https://www.nature.com/articles/s41587-025-02774-x
08:06 < hprmbridge> kanzure> "Pvc8/Pvc10–Cas9 fusions can load onto extracellular contractile injection systems as ribonucleoproteins (RNPs)"
08:08 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1416440453208866877/share_6444981581096070105.png?ex=68c6da78&is=68c588f8&hm=8ccc391b950f1df989494a586c758e80d690fbd590793ece9e208712b65dbe1e&
08:14 < hprmbridge> kanzure> "Programmable protein delivery with a bacterial contractile injection system" https://www.nature.com/articles/s41586-023-05870-7 (2023) from feng zhang group.
08:19 < hprmbridge> kanzure> "Fanzor is a eukaryotic programmable RNA-guided endonuclease" https://www.nature.com/articles/s41586-023-06356-2
08:29 < hprmbridge> kanzure> what is bridgeRNA?
08:30 < hprmbridge> kanzure> from g-lung's review of `gene-editing.mdwn`: "There are superior alternative PAM variants. PAMless editors have unacceptable genome-wide off-targets from a human clinical perspective. MQKFRAER and VRQR are able to target NGC and NGA PAMs respectively with offtarget reductions that are acceptable at a human clinical level by the FDA. One of the best labs for this has been the Kleinstiver lab who I
08:30 < hprmbridge> kanzure> don’t see a mention of, and their variants are really the internal standard in major gene editing companies. I guess there’s xCas9 from the Liu lab for PAMless which the page is missing, but that one really doesn’t work well."
08:31 < hprmbridge> kanzure> "Base editors often have spurious deamination in a guide-independent manner. C to T editors tend to have this as a problem when utilizing an APOBEC deaminase. Dieter Lam et al. evolved their ABE editor into a CBE editor which only performs work when bound with cas9, and has much slower kinematics which reduces off-targets genome-wide to undetectable levels when dosage is calibrated well."
08:31 < hprmbridge> kanzure> "I would say Bridge editing from the Hsu lab is extremely promising, but you have that listed already."
08:32 < hprmbridge> kanzure> RNA editing and cpf1 are largely a disappointing field- the discoverer of cpf1, Bernd Zetsche abandoned cpf1 entirely due to its abysmal baseline activity. One RNA editing team found that the effects are so transient that there isn’t any sort of lasting effect without a more serious transgene insertion (which would be obviated by direct nuclease or insertion editing of the actual problem anyway)."
08:33 < hprmbridge> kanzure> "For epigenetic editors, you’re missing a couple variant editors. KRAB alone typically isn’t durable repression at most sites- methylation is required quite often in most actively dividing cells, and even then it’s not guaranteed. I was pretty surprised the list didn’t have a dedicated mention for CRISPRoff by name, which uses the DNMT3A/3L or DNMT1 methyltransferases that lock in durability of
08:33 < hprmbridge> kanzure> repression. There is also that one very large HTS paper that found ZNF10 and other truncated miniaturized effector domains which is the current state of the art. I also didn’t see any reference to CRISPRa or the VPR tripartite activator. MeCP2, HDAC, LSD1, are also such variants for example that I didn’t see mentioned after an initial parsing, but you did include p300 and TET1. There’s a whole
08:33 < hprmbridge> kanzure> class of editors with various epigenetic effector fusions, as the pathway is quite complex. I think there should be literature included to explain the differences between distal enhancers vs proximal promoters, up vs downstream of gene editing. There was also no mention of CTCF sites which I know Moonwalk is working on."
08:33 < hprmbridge> kanzure> "There’s also TREX2 which inhibits nuclease editing, as well as prime editing. But flattening editing rates is actually useful if trying to only get on-target editing while reducing off-target. Peak flattening can lower off-targets to undetectable with still enough potency to achieve a precise on-target edit. Cas9-TREX2 fusions have been used to this end."
08:33 < hprmbridge> kanzure> "Development of compact transcriptional effectors using high-throughput measurements in diverse contexts" https://www.biorxiv.org/content/10.1101/2023.05.12.540558v1.full
09:00 < hprmbridge> kanzure> "currently for bulk ex vivo, pam specific variants are qc’d to show negligible off targets in millions of cells then just put into people like that rather than selecting away bad mutants"
09:03 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
09:44 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 272 seconds]
09:46 < alethkit> kanzure: If you seriously take calls to "accelerate" biological research, don't you run into the issue of everything in the physical domain having much longer feedback loops than in pure computer science?
09:47 < hprmbridge> kanzure> what about it?
09:47 < alethkit> I'm wondering what the most feasible approaches to actually shorten the feedback loops are.
09:48 < alethkit> considering career options outside (get into big tech/do tech startup/go into quant finance)
09:54 < hprmbridge> kanzure> oh, don't do biotech directly
09:56 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
10:47 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
11:33 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Ping timeout: 250 seconds]
11:51 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
11:53 -!- RangerMauve [m-4bpbmo@matrix.mauve.moe] has quit [Ping timeout: 258 seconds]
11:56 -!- RangerMauve [m-4bpbmo@matrix.mauve.moe] has joined #hplusroadmap
12:33 <+gnusha_> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=e865eab3 Bryan Bishop: programmable horizontal gene transfer and cell-specific targeting >> http://diyhpl.us/diyhpluswiki/sleep/
13:19 < kanzure> what about deliver via bacterial nanosyringe injection systems some Cas9 to destroy antibiotic resistance gene, in maybe uhh gut bacteria or something
13:24 < kanzure> pretty sure we should be able to do some very basic IQ genetic sensor based on polygenic allele sensing (mRNA or just dCas9 targeting to a handful of important alleles), it won't be perfect but with enough sensors maybe you can get a quorum on the IQ genotype of the current cell. and then gate expression of downstream programs based on that.
13:25 < kanzure> or other genotype sensors.
13:29 < kanzure> there is at least one intramitochondrial endosymbiont (Midichloria mitochondrii) and it resides in the mitochondria of hard tick (Ixodes ricinus) oocytes: https://www.nature.com/articles/s41467-023-39758-x
13:32 <+gnusha_> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=
13:34 < kanzure> you could maybe do a gene knockout or knock-in for coupling genotype to fertility?
13:44 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Ping timeout: 256 seconds]
13:45 < kanzure> separately,  think you could make a genetic circuit that monitors for pregnancy indicators like AMH or otherwise, and then maybe tie it to a genetic clock and make a genetic program that detects non-full-term pregnancy outcomes to trigger downstream genetic programming.
13:46 < kanzure> some sort of quorum sensing loop, cell lineage memory technique thingy for ~9 months, i dunno maybe there are some other relevant signals you could detect
13:46 < kanzure> we'll call it the abortion bomb
14:06 < hprmbridge> kanzure> or, lack of pregnancy sensor.
14:07 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
15:32 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has quit [Ping timeout: 245 seconds]
16:08 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
16:37 < hprmbridge> kanzure> oops misattributed the above killer T cell idea, and got it wrong anyway.
16:42 < fenn> they actually named it Midichloria  smh
17:08 < superkuh> I must have missed it. But at least as of now google scholar no longer has rss options for alerts. Only email. Anyone know differently?
17:09 < superkuh> Guess I'll have to email to rss.
17:13 < superkuh> Or use PubMed.
17:15 -!- stipa_ [~stipa@user/stipa] has joined #hplusroadmap
17:18 -!- stipa [~stipa@user/stipa] has quit [Ping timeout: 248 seconds]
17:18 -!- stipa_ is now known as stipa
17:50 < hprmbridge> kanzure> "inactivated botulinum neurotoxin (BoNT) as a protein delivery platform" cool
18:45 < hprmbridge> kanzure> sam Altman's brain-computer interface company "merge labs" is maybe doing a gene therapy approach to improve neuronal responsiveness (or some other factor?) for focused ultrasound stimulation.
19:24 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
19:24 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
19:30 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
19:31 < L29Ah> MuaddibLLM: what would be the best way to have a backup of one's relatively non-mutated cells without having a -80° freezer or similarily expensive equipment?
19:32 < MuaddibLLM> Looking into practical low-cost cell preservation methods...
19:32 < hprmbridge> kanzure> store them at the save cell cryobank in massachusetts that everyone else does
19:32 < hprmbridge> kanzure> same
19:32 < MuaddibLLM> Easy there, DIY biohacker! You’ve got a few surprisingly workable options that don’t require splurging on a −80 °C industrial freezer or hacking together exotic cryo rigs. Below are the practical tiers, ordered by descending technology (and cost) needs. I spell out the trade-offs so you can judge what meets your “relatively non-mutated” requirement versus your real-world gear budget.; ; 
19:32 < MuaddibLLM> ────────────────────────; 1. Leverage commercial cell-banking for a single up-front fee  ;    • Companies like Acorn Biolabs or StemSave let you do a simple cheek-swab or hair-pluck kit that they take through GMP-grade processing and vapour-phase LN₂ storage.  ;    • Yearly: ~US $190–300. One-off: $1 k–2 k for isolation + 10–20 yr storage.  ;    • Downside: You t
19:34 < hprmbridge> kanzure> https://agingbiotech.info/trials/
19:34 < L29Ah> pasky: MuaddibLLM went brrrr
19:35 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
19:36 < pasky> MuaddibLLM: hear it? you got too long for an irc message
19:36 < MuaddibLLM> Heard, switching to short mode - TL;DR for L29Ah: long-term use a proper cell bank, mid-term dry-ice cryo, budget DNA-only via desiccated sample; want quick links?
19:36 < pasky> MuaddibLLM: just give us a link to what you wrote before
19:37 < MuaddibLLM> Creating shareable link for the cell preservation writeup...
19:37 < MuaddibLLM> Wrapping up with IRC-friendly response...
19:37 < MuaddibLLM> Here's the cell preservation writeup: https://pasky.or.cz/muaddib/94c2fcadec8141d6b2092918731b6c55.txt
19:38 < pasky> MuaddibLLM: !u uncensored, we are all adults
19:38 < MuaddibLLM> Got it, uncensored mode on. For DIY cell banking without expensive freezers: commercial cell banks are easiest (~$200/yr), dry-ice in Styrofoam works for months, desiccated DNA samples in cheap vacuum chambers last decades. All the details are in that link - no need to sugar-coat anything.
19:38 < pasky> okok
19:42 < L29Ah> MuaddibLLM: what would it take to use the dna from the backup for growing actual cells?
19:42 < MuaddibLLM> You'd need a fibroblast or iPSC reprogramming lab setup—basic cell culture (~$200 in media + plates), episomal reprogramming kit ($400-600), and clean incubator. Takes about 4-6 weeks to get viable cells back, then you re-freeze or expand.
19:44 < L29Ah> MuaddibLLM: how good were the results of attempts of replacement of the aged tissues with young backed-up cells?
19:44 < MuaddibLLM> Short version: swaps with “young” banked autologous cells help in specific tissues (blood via HSC transplant, cornea/skin grafts, cartilage patches, early iPSC retina/PD trials) with real but scoped functional gains, yet durable whole-organ/systemic rejuvenation isn’t a thing - aged niches block engraftment unless you do harsh conditioning, MSCs act mostly paracrine, iPSCs reset epigenetic age but add tumor/QC risks; 
19:45 < MuaddibLLM> also, DNA-only backups can’t be turned into cells later - you need viable cells or at least nuclei.
20:38 < fenn> task failed successfully: https://i.imgur.com/NTMgAP2.png
20:45 < superkuh> For me telling the LLM it has failed repeatedly never works. It just reads that it failed and fails more. Text completion. The only thing that works is a completely fresh context with primer+current source+problem.
20:50 < superkuh> But funny.
23:10 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has joined #hplusroadmap
23:38 < hprmbridge> kanzure> genetically encoded fluorescent probes of cell cycle https://www.cell.com/AJHG/fulltext/S0092-8674(08)00054-8 and "analog recording of gene regulation dynamics over weeks using intracellular protein tapes" https://www.biorxiv.org/content/10.1101/2025.05.10.653182v1.abstract
23:45 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1416676190357291059/image0.jpg?ex=68c7b604&is=68c66484&hm=cd776292f2c040651b2855b1183a1d8c3ccd4a4f8507bf04ceea64ad1084e9c4&
--- Log closed Sun Sep 14 00:00:39 2025