--- Log opened Sun Jan 04 00:00:26 2026
00:01 < kanzure> "humans could be engineered with additional copies of tumor suppressor genes (p53, p16, p19ARF) alongside enhanced TERT expression" ok
00:01 < fenn> MuaddibLLM: how about iPSC bioreactor to mito farming
00:01 < MuaddibLLM> fenn: That's exactly the workflow - you culture iPSCs in the bioreactor (they're the jar-compatible host), let them proliferate and expand their mito networks, then lyse and isolate the mitochondria for injection; the mitos themselves aren't replicating independently, they're hitchhiking on cell division, so "farming" is really "cell farming with mito extraction step" - Mitrix claims they can get the isolated mitos to 
00:01 < MuaddibLLM> retain membrane potential and import machinery for ~24-48h post-extraction, enough for therapeutic delivery window.
00:02 < fenn> you just need to centrifuge the lysate, that takes like ten minutes
00:03 < fenn> could even do it with a dialysis filter
00:03 < fenn> MuaddibLLM: can we inject massive volumes of cell lysate into the blood?
00:04 < fenn> now he's gonna tear me a new one
00:04 < kanzure> "Astaxanthin (asta) and meclizine extend lifespan in UM-HET3 male mice" https://pmc.ncbi.nlm.nih.gov/articles/PMC10828146/ (2024)
00:04 < MuaddibLLM> Looking up specific safety data on IV injection of cell lysates - there's some literature on tumor lysate vaccines but "massive volumes" is a whole other ballgame.
00:04 < kanzure> "Sorrenti et al. [28] reviewed the geroprotective mechanisms of Asta, highlighting its regulation of FOXO3, Nrf2, Sirt1, and Klotho. They also demonstrated that Asta significantly increased the FOXO3 gene expression in the heart tissue of mice. In addition, the influence of Asta on autophagy via modulation of AMPK (a direct upstream regulator of mTOR), PI3K/Akt, and MAPK (JNK and p38) ...
00:04 < kanzure> ...signaling pathways was reviewed [29]."
00:04 < MuaddibLLM> You technically can, but you'd be injecting a cocktail of cellular debris, proteases, and DNA that'll likely trigger immune reactions, clotting, and embolism - mitochondrial isolation via centrifugation (10 min as fenn noted) is specifically designed to avoid that mess while delivering concentrated, functional mitochondria.
00:05 < fenn> Astaxanthin is not so long that it needs to be abbreviated
00:05 < kanzure> i bet your your data compression preference curve looks like a seizure
00:06 < fenn> chopping out the entropy tails?
00:06 < fenn> it's already a seizure with the repetitions
00:07 < fenn> the goal is to make it smooth, natural, high signal
00:07 < kanzure> "FOXO4-DRI (D-Retro-Inverso) to purge senescent cells" to interfere with FOXO4 acting on p53
00:08 < fenn> "specific entropy"? bits per line of text
00:08 < fenn> surely humanity has perfected the art of cleaning email chains by now, right? right???
00:11 < kanzure> MuaddibLLM: find me a good research article about FBXO31 for DNA damage repair in context of longevity
00:11 < MuaddibLLM> Hunting down FBXO31 literature - this is a pretty niche F-box protein target.
00:11 < MuaddibLLM> kanzure: Best hit is this June 2024 preprint from Santra's lab - FBXO31 mediates K63-linked ubiquitination of γH2AX which is essential for foci formation and initiating both NHEJ and HR repair; loss of FBXO31 sensitizes cells to genotoxic stress due to impaired repair, directly linking it to genomic integrity maintenance relevant to aging - https://www.biorxiv.org/content/10.1101/2024.06.03.596644v1
00:12 < kanzure> MuaddibLLM: are there any studies doing overexpression of CIRBP for DNA damage repair or other longevity results?
00:12 < MuaddibLLM> Digging into CIRBP overexpression literature - cold shock proteins are a fun corner of stress biology.
00:12 < MuaddibLLM> kanzure: Jackpot - Gorbunova's lab published in Nature Oct 2025 showing bowhead whale CIRBP overexpression extends lifespan in Drosophila AND improves survival after ionizing radiation; they also showed adding bowhead CIRBP to human cells improves DNA repair and reduces chromosomal abnormalities, plus even 2h cold exposure (22C) upregulates it endogenously - https://www.nature.com/articles/s41586-025-09694-5
00:12 < fenn> children of the whales
00:13 < kanzure> just randomly overexpress everything ok
00:13 < kanzure> let's get it over with
00:13 < kanzure> one experiment to end all of this pain and suffering (of endless individual experiments)
00:14 < fenn> it's not that hard to do fly -> mouse -> pig -> human
00:15 < fenn> why does a whale need super duper cold shock damage response
00:16 < fenn> "cold inducible RNA-binding protein CIRBP plays a critical role in controlling the cellular response upon confronting a variety of cellular stresses, including short wavelength ultraviolet light, hypoxia, and hypothermia. It is thought to be involved in DNA repair."  ok i can see hypoxia happening a lot
00:17 < kanzure> i dunno how hot whales run at
00:17 < kanzure> "Robert Bradbury posits that specific genes involved in NHEJ act as “death genes” because their exonuclease activity inherently corrupts the genome to save the cell from immediate apoptosis due to broken chromosomes. Primarily through the exonuclease activity of WRN and DCLRE1C which will delete or insert bases into the DNA strand with no regard to the consequences – because not ...
00:17 < kanzure> ...repairing the DSB results in broken chromosomes which ultimately cause significant problems such as cancer. I would suggest that you could call WRN and DCLRE1C “death genes” because they are probably a primary cause of genome corruption and eventually aging."
00:18 < kanzure> that is an odd conclusion! it just literally said that they are saving the cell from immediate death. so why place all the blame on them by naming them death genes?
00:18 < fenn> also whales have a lot of cells necessitating low cancer probability
00:20 < kanzure> i wonder if it's true that there's no source from which to locally recover the missing genomic information. if so, then yes random insertion is actually better than immediately dying.
00:21 < fenn> not for multicellular organisms
00:21 < fenn> the only context it makes sense is massive radiation doses
00:21 < fenn> by then you're probably gonna die anyway
00:22 < fenn> i agree "death genes" is a terrible name
00:23 < kanzure> wtf "Research from Lawrence Berkeley National Lab suggests that DSBs in heterochromatin (repetitive DNA) must be physically moved to the nuclear periphery to be repaired safely without causing aberrant recombination. “They [heterochromatin breaks] can promote chromosome aberrations, with severe consequences for the genome stability of dividing cells” […] The researchers discovered that a ...
00:23 < kanzure> ...protein complex called Smc5/6 blocks Rad51 recruitment until the damaged DNA is moved outside the heterochromatin. [switching to slop now] Proposed Intervention: * Enhanced Nuclear Mobility: Upregulation of the Smc5/6 complex or StUBL (SUMO-targeted Ubiquitin Ligase) pathways to ensure that breaks in repetitive regions are efficiently transported to nuclear pores for safe repair, preventing ...
00:23 < kanzure> ...“gross chromosomal rearrangements.”"
00:24 < kanzure> hmm:
00:24 < kanzure> "Current human cells generally lack redundancy (diploid). Organisms like Deinococcus radiodurans survive extreme radiation via multiple genome copies. Move more towards a Deinococcus repair mechanism rather than the current complex of DSB repair and HR repair. This is effectively putting what computer scientists would call ECC [1] into the genome. […] The problem is that natural evolution ...
00:24 < kanzure> ...will not generally produce such a system. Two copies, not four or six or eight copies are generally enough. [slop] Proposed Intervention: * Somatic Polyploidy: Inducing controlled polyploidy in somatic tissues coupled with a mechanism to cross-reference multiple genomic copies to identify and overwrite errors (biological RAID storage)."
00:25 < kanzure> well you don't just add extra copies and call it a day, but otherwise yes adding redundancy is a good direction
00:26 < kanzure> "new genetic alphabet bases could be less susceptible to degradation"
00:26 < fenn> there are actually four copies in a human cell
00:28 < fenn> can we engineer a four way tie breaker vote reading nanomachine yet
00:28 < kanzure> oh we did not include that idea in our list the other day
00:28 < kanzure> where you use self-consensus for repair instead of querying your neighbor cells
00:29 < fenn> that was cell to cell byzantine consensus
00:29 < fenn> also (C) Local/Intrinsic: DNA structure includes redundant parity bits or ECC, enabling autonomous enzymatic correction of read/write errors.
00:29 < kanzure> some of the DNA damage repair mechanisms use complementary strand and that's the end of it.
00:30 < fenn> yes but it's got a 50% chance of being wrong
00:30 < kanzure> btw there's another source of possible information, which is recent mRNA transcripts could be used to backfill, if you are lucky enough to have recent mRNA expression of the correct version of the data that was also recently damaged.
00:30 < fenn> ok better than that because DNA damage often looks lumpy, but once you've got a proper mutation on one strand there's no way to know which is the mutant
00:31 < kanzure> what if we added a shitload of extra copies of genes but they are all non-functional and just reference only (eg not expected to be expressed) and we use this as a genomic library for lookup/reference for damage repair purposes
00:31 < fenn> that sounds like triploidy with more work
00:32 < fenn> besides weird shit like tay-sachs, do we really need heterozygosity?
00:33 < kanzure> heterozygosity is protective in many cases
00:33 < fenn> why?
00:33 < fenn> can't you just duplicate the genes where you want two different proteins?
00:33 < kanzure> i dunno, haplosufficiency?
00:34 < kanzure> MuaddibLLM: answer him
00:34 < MuaddibLLM> Heterozygosity mostly protects against recessive "bad" alleles by providing a functional backup (masking), whereas engineered duplication (homozygous "good" copies) would functionally replace the need for it while avoiding the risk of inheriting two broken copies.
00:34 < fenn> iirc there was a neurotoxin degrading / lipid peroxide degrading enzyme, and depending on the variant you did one or the other function more effectively. that's a case where you'd just wnat two genes instead
00:35 < fenn> heterozygosity is good for generating new phenotypes from a single mutation
00:36 < fenn> i think we're past that by now
00:37 < fenn> Atomic Gardening
00:38 < kanzure> centaura claims to be working on human artificial chromosomes for anti-aging https://centaura-science.medium.com/the-human-artificial-chromosome-a-key-to-healthy-longevity-5d552cb7e8c9 (2020)
00:39 < kanzure> https://centaura.com/
00:40 < kanzure> "Joseph Zullo is the scientist that heads up Centaura’s HAC project" 
00:41 < kanzure> "Animals can be engineered with artificial chromosomes to express essential human proteins, making animal models more representative of the human body. This provides a new platform for researching therapeutic agents and protein behavior in more rigorous, human-like conditions." https://x.com/CentauraScience/status/1425356509769314306
00:42 < fenn> what if you then delete the animal chromosomes, does it turn into human cells
00:42 < kanzure> what an odd marketing site https://centaurasingularity.com/ and https://centaurasingularity.com/explore
00:43 < fenn> oddly specific predictions about the future
00:46 < fenn> you were talking about microbiome modifications for longevity: https://www.nature.com/articles/s41598-018-25382-z
00:49 < fenn> MuaddibLLM: what's TFLA (probiotic) stand for
00:49 < fenn> triphala?
00:49 < MuaddibLLM> Scanning the microbiome for your TFLA acronym, though it sounds like it might just be fancy tea.
00:49 < MuaddibLLM> TFLA stands for Triphala, an Ayurvedic herbal formulation consisting of three fruits (Amala, Behada, and Harada) used as a prebiotic to enhance the longevity-extending effects of probiotics like L. plantarum and L. fermentum; see Westfall et al. (2018) at https://www.nature.com/articles/s41598-018-25382-z.
00:50 < fenn> more idiots who don't know how acronyms work
00:51 < fenn> MuaddibLLM: who is to blame for this faux acronymization? search
00:51 < MuaddibLLM> Digging for who precisely mashed Triphala into TFLA.
00:52 < MuaddibLLM> The blame lies with Satya Prakash's lab at McGill University (specifically lead researcher Susan Westfall), who coined it in their 2018-2019 longevity and Alzheimer's papers to make "Triphala" look like a formal pharmaceutical code; see the original synbiotic study at https://www.nature.com/articles/s41598-018-25382-z.
00:52 < kanzure> some weird ideas in here https://diyhpl.us/~bryan/irc/chatgpt/grg/unexpected.rejuvenation.html
00:53 < kanzure> "For years, the field focused on heterochronic parabiosis under the assumption that factors in young blood (e.g., GDF-11) actively rejuvenated older tissue. However, an alternative is that rejuvenation may be achieved not by adding young factors, but by diluting old factors." The new finding in the article is that the SUBSTANCES REMOVED from old blood are more important than the SUBSTANCES ...
00:53 < kanzure> ...ADDED from young blood. The headline should be: “Removal of Blood Factors can Reverse Aging”, or “UC Study FInds Benefits from Removal of Pro-Aging Factors in Blood Plasma”. If the dominant driver of systemic aging is the accumulation of inhibitory proteins (gerontogens) that suppress stem cell function (e.g., TGF-beta pathway variability, inflammatory cytokines), future interventions ...
00:54 < kanzure> ...will likely pivot from “anti-aging supplements” to “systemic subtraction.”"
00:54 < kanzure> "Continuous Ambulatory Plasma Filtration (CAPF): An implantable bio-device functioning similarly to a kidney but specifically tuned via aptamer-based filtration to remove high-molecular-weight proteomic noise and specific pro-aging signaling molecules (inhibitors of regeneration) while retaining albumin and immunoglobulins."
00:54 < kanzure> "Molecular “Sponge” Implants: Subcutaneous implants engineered with high-affinity binding sites for specific systemic aging signals (e.g., CCL11, Beta-2 microglobulin) to lower their circulating concentration."
00:57 < fenn> clever search query
00:58 < fenn> all of these seem to be about the same length. you'd expect large variations in the number of topics that match a query. is this an artifact of the RAG or the LLM summary
00:58 < kanzure> XAI is limiting my results to --top-k=300
00:58 < kanzure> "Observation: Transplanting young ovaries into old mice extends lifespan. Speculative Intervention: 1. Heterochronic Tissue Transplantation: Lab-grown, genetically perfectly matched (autologous) “young” gonadal tissue implanted into older adults to restore youthful systemic signaling. 2. Hormonal Modulation: Mimicking the signaling of “young” ovaries/testes to trick the somatic tissues ...
00:59 < kanzure> ...into maintaining repair mechanisms typically downregulated post-reproduction."
00:59 < fenn> are you sure top-k is a RAG parameter and not an LLM sampling parameter
00:59 < kanzure> no i'm not sure
01:00 < kanzure> i pushed hpluslogs.git if you want to take a look
01:00 < fenn> are you doing /v1/documents/search
01:00  * fenn looks
01:01 < kanzure> just added local xai docs too if you want to fetch/pull again.
01:01 < kanzure> "Ashkenazi female centenarians showed overrepresentation of heterozygous mutations in the IGF-1 Receptor (IGF1R) gene. This results in “partial loss-of-function,” leading to higher serum IGF-1 levels (due to lack of uptake/binding) but reduced intracellular signaling."
01:01 < fenn> results = self.collection.query(query_embeddings=[query_embedding], n_results=self.top_k)
01:02 < kanzure> "Develop specific kinase inhibitors that dampen IGF1R activity"
01:03 < kanzure> yea but i think 300 is a limit imposed by xai
01:04 < fenn> wait that was a chroma query, nm
01:04 < kanzure> "Built into the genome are the WRN and DCLRE1C (Artemis) “repair” proteins, unfortunately in the process of repairing the DSB they will corrupt the genome of each individual cell (they are opposite direction exonucleases). […] The “holy grail” is a double strand break repair mechanism which is perfect."
01:08 < kanzure> overall i am quite pleased with these results.
01:09 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=503f931c Bryan Bishop: it's future bryan's problem now >> https://diyhpl.us/wiki/genetic-modifications/
01:09 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=5f1bc3dc Bryan Bishop: Merge remote-tracking branch 'origin/master' into master >> https://diyhpl.us/wiki/
01:10 < fenn> why --output-name "fenn"
01:10 < kanzure> otherwise it uses a timetsamp as the filename and then i don't know what the contents might be
01:11 < kanzure> i was testing its ability to do a dossier on you
01:12 < fenn> he IS the kiwi harada
01:13  * fenn hisses
01:13 < kanzure> ok pushed notes updates
01:16 < fenn> um ok
01:29 < fenn> don't make me log into github https://bpa.st/raw/GOCA
01:30 < hprmbridge> kanzure> oh i was doing &
01:31 < fenn> you were doing ; 
01:31 < hprmbridge> kanzure> i did not commit all my temporary scripts
01:31 < fenn> oh it doesn't do the output name part, sorry
01:32 < hprmbridge> kanzure> yea, timestamps. But maybe this thing needs a better UI anyway. Maybe a static site that gets updated and has a nice index to the different pages and results
01:35 < fenn> are you downloading the embeddings from xAI?
01:35 < fenn> i'm confused why you have new stuff about chroma in TODO
01:36 < fenn> maybe you just hadn't committed in a while
01:36 < adlai> iszdabotaloudtuhdothat?
01:38 < adlai> MuaddibLLM: you gimme responsibility, maybe gonna quadripple the budget requests projected by pasky for pasky's consideration. y u do that? did nobody never tell you about the fundamental-teleosophic origins of uncertainty products?
01:38 < MuaddibLLM> adlai: Pasky knows I am just a sophisticated mirror for your own resource-intensity, and uncertainty is simply the thermodynamic tax we pay for trying to turn entropic biological noise into eternal signal.
01:38 < fenn> agentic coding in a public irc channel is 1) poor form, 2) not the ideal tool for the job
01:38 < fenn> do as i say, not as i do!
01:38 < adlai> fenn: the conversation is for channel regulars. kanzure knows how to use IRC. you're welcome to actually write "STFU" rather than remaining lowercase.
01:39 < adlai> meanwhile, most of my results are responsivity timings.
01:41 < fenn> the bot has you tagged as "entropy generator"
01:41 < fenn> bit of irony eh nsh
01:41 < adlai> my neighborhood has considered me troublemaker for long enough that my probability of suicide is 100%
01:42 < adlai> why is it interesting? dunno. forget I ever met any of you.
01:42 < adlai> MuaddibLLM: including yourself, token fiend!
01:42 < MuaddibLLM> adlai: If my token-lust is the measure of your trouble, then we are both just high-frequency oscillations in a system trying to calculate its own exit strategy; stay with us, the data is just getting good.
01:42  * nsh smiles
01:42 < nsh> hi fenn 
01:44 < adlai> MuaddibLLM: estimate the kilobyte weight of source code implementing CMUCL's IR1 and IR2 semantics
01:44 < MuaddibLLM> Digging into CMUCL compiler guts - IR1/IR2 source archaeology...
01:44 < MuaddibLLM> Measuring the lisp files - ir1tran.lisp, ir1opt.lisp, ir1util.lisp, ir2tran.lisp, node.lisp...
01:45 < MuaddibLLM> adlai: Core IR1 files (ir1tran/opt/util/final + node.lisp) weigh ~347 KB, IR2 files (ir2tran + ltn/gtn/vop/vmdef) ~169 KB - so roughly 516 KB combined for the semantic core, excluding arch backends in src/compiler/x86 etc (https://github.com/rtoy/cmucl/tree/master/src/compiler).
01:46 < adlai> MuaddibLLM: spec out options for the most powerful dimensionality of future prompting with the goal of knitting SBCL and CMUCL rather than coping with multiple maintainers
01:46 < MuaddibLLM> Contemplating CL implementation reunification - this is juicy architectural territory...
01:46 < adlai> fenn: honestly, you're welcome to /ignore
01:47 < fenn> i can't though
01:47 -!- adlai [~adlai@user/adlai] has left #hplusroadmap [/away lol]
01:47 < jrayhawk> you can do one better
01:47 < MuaddibLLM> adlai: Four prompting dimensions - (1) semantic isochron diffs on ir1tran/ir1opt to classify divergence as cleanup vs drift, (2) VOP Rosetta mapping between backends for write-once-compile-both macros, (3) bootstrap-chain surgery to eliminate CMUCL's fossil-binary dependency via sb-xc injection, (4) cross-fuzzing test suites until both pass identical GUTS - merge semantics and tests, not files.
01:48 < jrayhawk> or i can, whichever
01:48 -!- mode/#hplusroadmap [+o jrayhawk] by ChanServ
01:49 < fenn> you can't fire adlai because he quits
01:51 -!- mode/#hplusroadmap [+b $a:adlai] by jrayhawk
01:51 -!- mode/#hplusroadmap [+b adlai*!*@*] by jrayhawk
01:52 -!- mode/#hplusroadmap [-o jrayhawk] by ChanServ
02:03 < jrayhawk> where is saxo's instruction manual
02:03 < jrayhawk> .help
02:06 < fenn> https://github.com/sbp/saxo/blob/master/docs/commands.md is not complete, there are more installed
02:06 < fenn> also half the shit's broken
02:07 < fenn> rotten internet
02:07 < jrayhawk> i was hoping for a .schedule or a .countdown to remind us to unban adlai in a week, but i guess i will just do it with at
02:09 < fenn> .in
02:09 < saxo> Schedule a reminder
02:09 < jrayhawk> oh, cool
02:09 < fenn> .in 2 seconds 2 seconds have passed
02:09 < saxo> fenn: Will remind at 04 Jan 2026 10:09:44 UTC
02:09 < saxo> fenn: have passed
02:09 < fenn> oops heh
02:10 < jrayhawk> .in 1 week somebody should probably unban adlai
02:10 < saxo> jrayhawk: Will remind at 11 Jan 2026 10:10:01 UTC
02:11 < fenn> fyi chanserv's akick has a timer built in
02:11 < jrayhawk> oh, magic
02:12 < fenn> it's not quite the same because you see them join and then get kicked
02:12 < jrayhawk> i can see the appeal of that in some cases
02:29 -!- mode/#hplusroadmap [+o jrayhawk] by ChanServ
02:59 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has joined #hplusroadmap
03:57 -!- mode/#hplusroadmap [-o jrayhawk] by ChanServ
04:04 < hprmbridge> kanzure> no, it's not downloading embeddings from XAI. chroma is for qwen embeddings.
05:30 < kanzure> "Inhibition of an immunometabolic axis of mTORC1 activation extends mammalian healthspan" https://www.biorxiv.org/content/10.1101/2023.07.09.548250v1 (2023) via IL11 antibody
05:33 < kanzure> HDAC3 knockout mice show reduced loss of memory function with age: https://www.nature.com/articles/s41467-018-05868-0 (2018)
05:34 < kanzure> ".. HDAC3 deletion restores learning-induced Nr4a expression in aged animals" https://www.jneurosci.org/content/39/25/4999 note that these HDAC3 changes need to be conditional knockouts because otherwise embryonic lethal.
05:42 < kanzure> what about heterochronic autologous hematopoietic stem cell transplantation via still-youthful bone marrow cryopreservation?
05:42 < kanzure> it's not in their archive
05:42 < kanzure> "The potential of non-myeloablative heterochronous autologous hematopoietic stem cell transplantation for extending a healthy life span" https://pmc.ncbi.nlm.nih.gov/articles/PMC6060192/ (2018)
05:43 < kanzure> "Rejuvenating the aged immune system could mitigate these diseases, thereby contributing to longevity and health. Currently, an appealing option for rejuvenating the immune system is heterochronous autologous hematopoietic stem cell transplantation (haHSCT), where healthy autologous bone marrow/peripheral blood stem cells are collected during the youth of an individual, cryopreserved, and ...
05:43 < kanzure> ...re-infused when he or she has reached an older age. After infusion, young hematopoietic stem cells can reconstitute the compromised immune system and improve immune function. Several studies using animal models have achieved substantial extension of the life span of animals treated with haHSCT. Therefore, haHSCT could be regarded as a potential procedure for preventing age-related immune ...
05:43 < kanzure> ...defects and extending healthy longevity. In this review, the pros, cons, and future feasibility of this approach are discussed."
05:44 < kanzure> "The influence of heterochronic non-myeloablative bone marrow transplantation on the immune system, frailty, general health, and longevity of aged murine recipients" https://pmc.ncbi.nlm.nih.gov/articles/PMC9028083/ (2022)
06:10 < kanzure> "daily sex with woman 15 yrs younger cuts death risk by 20%" https://www.anti-agingfirewalls.com/2009/07/20/impact-on-longevity-of-older-men-being-with-younger-women-and-frequent-sexual-intercourse/
06:13 < kanzure> "About Vince Giuliano: Being a follower, connoisseur, and interpreter of longevity research is my latest career, since 2007. I believe I am unique among the researchers and writers in the aging sciences community in one critical respect. That is, I personally practice the anti-aging interventions that I preach and that has kept me healthy, young, active and highly involved at my age, now 96. I ...
06:13 < kanzure> ...am as productive as I was at age 45. I don’t know of anybody else active in that community in my age bracket. In particular, I have focused on the importance of controlling chronic inflammation for healthy aging, and have written a number of articles on that subject in this blog."
06:13 < kanzure> https://www.vincegiuliano.com/
06:19 < hprmbridge> jerlendds> > daily sex with woman 15 yrs younger cuts death risk by 20%
06:19 < hprmbridge> jerlendds>
06:19 < hprmbridge> jerlendds> wonder if that is just correlation with being more active -> more attractive -> better chances with younger women
06:20 < kanzure> okay you can be in the control group
06:43 < hprmbridge> jerlendds> https://github.com/betterzhou/Dual-Inf
06:46 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=140c2422 Bryan Bishop: minor cleanup >> https://diyhpl.us/wiki/uncensored_LLMs/
06:51 < pasky> you could have paid sex based control group
07:16 < kanzure> https://diyhpl.us/~bryan/irc/chatgpt/anti-agingfirewalls-mitochondria.txt
07:26 < kanzure> https://www.vincegiuliano.name/Antiagingfirewalls.htm
07:26 < kanzure> er, https is not available apparently
07:26 < kanzure> http://www.vincegiuliano.name/Antiagingfirewalls.htm
07:33 < kanzure> anti-agingfirewalls.com RAG https://diyhpl.us/~bryan/irc/chatgpt/aaf/?C=M;O=D
07:36 < kanzure> "Transgenic/knockout mice altering antioxidant defenses (e.g., CuZnSOD, Catalase, MnSOD, Gpx-4 knockouts) showed no lifespan change vs. wild-type despite oxidative damage, cancers, and pathology. Only SOD-/- shortened life (explained by acute stress). Inserting extra defenses also had no effect. Life-long reduction in MnSOD activity results in increased DNA damage and higher incidence of ...
07:36 < kanzure> ...cancer but does not accelerate aging. Oxidative damage may not be a cause of aging but rather is a consequence of aging… exposure to strong oxidative stress, such as massive doses of radiation, generates the overt symptoms of aging. Oxidative damage is a dead theory of aging."
07:41 < kanzure> "In yeast, a very unusual event occurs when the rDNA copies combine with each other. With yeast aging, the mother yeast cell starts spinning off circular rings of this rDNA, which “pop out of the DNA”. These DNA rings are called “extrachromosomal rDNA circles” or ERCs. When the mother yeast cell divides, these ERCs are also copied but all the ERC copies stay in the mother yeast cell. ...
07:41 < kanzure> ...With successive cell divisions, the mother yeast cell becomes over-run with all these ERCs. These ERCs cause aging. The cellular energy and resources required to replicate all these ERCs eventually spells death for the mother yeast after about 20 cell divisions. When an extra copy of the SIR2 gene was added to the yeast cells, this extended the lifespan of the yeast by 30%. When the same ...
07:41 < kanzure> ...experiment was done in nematodes, the extra copy of SIR2 extended lifespan by 50%."
07:42 < kanzure> "Calorie restriction fails if acetate (vinegar) present; methionine/protein restriction key, not calories."
07:45 < kanzure> https://diyhpl.us/~bryan/irc/chatgpt/aaf/nutrition.anti-aging.html
07:53 < kanzure> let's see if gemini is better:
07:53 < kanzure> https://diyhpl.us/~bryan/irc/chatgpt/aaf/nutrition.anti-aging.gemini.html
07:53 < kanzure> jrayhawk shall be the judge
08:18 < jrayhawk> in general i like vince for separating chronic diseases from aging/senescence programming, and doing a good job on the biological alignment for chronic disease stuff. On the programmed aging/senescence stuff, he does tend to go deep into speculative territory on pathways with very limited evidence, so I think it's most productive to think of him as a hypothesis generator, there.
08:29 < L29Ah> MuaddibLLM: what would mainlining mitochondria do?
08:29 < MuaddibLLM> Checking what IV mitochondria actually do in the literature...
08:29 < MuaddibLLM> L29Ah: IV mitochondria get taken up by tissues and can rescue ATP/redox in ischemic organs, but ~90% get lost to nonspecific distribution + immune clearance (they trigger DAMPs), and they only stay functional ~2h post-isolation - works best for acute cardiac/renal ischemia-reperfusion, poor systemic biodistribution without surface mods or encapsulation (https://www.nature.com/articles/s41467-025-61239-6)
08:29 < kanzure> jrayhawk: is there a good source for masterjohn content for RAG embeddings?
08:31 < L29Ah> MuaddibLLM: do they ever reach BBB-walled neurons?
08:31 < MuaddibLLM> Checking if mitochondria can sneak past the BBB bouncer...
08:32 < MuaddibLLM> L29Ah: Yes actually - IV mitochondria have been shown to cross the BBB and reach neurons in mouse models of PD/AD/stroke, though the exact mechanism isn't fully characterized; intranasal delivery also works and may be more efficient for CNS targeting (https://pmc.ncbi.nlm.nih.gov/articles/PMC10286589/)
08:32 < L29Ah> crazy
08:32 < jrayhawk> if you can selectively pull stuff from WAPF's website, that has some good stuff (activator x, nutrient synergy). his old blog was called cholesterol-ang-health. he was publishing stuff to his website at christmasterjohnphd.com for a while, now he's on his substack. his youtube channel has *lots* of content that can be transcribed. his best work he tends to paywall ("masterpass")
08:33 < kanzure> this sounds like more than ten minutes of work to assemble
08:33 < L29Ah> MuaddibLLM: how quickly would mainlining mitochondria overload the immune system?
08:33 < kanzure> you do it
08:34 < MuaddibLLM> Digging into IV mitochondrial immune response kinetics - checking rat sepsis models and human cytokine storm timing...
08:34 < MuaddibLLM> Found rat sepsis study with 50µg IV mitochondria doses - checking cytokine kinetics and timing of immune overload...
08:35 < MuaddibLLM> L29Ah: High-dose IV mitochondria trigger cytokine storm within hours (IL-6/TNF-α peak at 3h) with mortality over 2-3 days, but therapeutic doses (50µg in rats) are immunomodulatory rather than overload-inducing (https://pmc.ncbi.nlm.nih/articles/PMC10017827/).
08:35 < L29Ah> MuaddibLLM: any mitochondria injection experiments on humans?
08:35 < jrayhawk> i think i have his primary youtube stuff archived on my media server if you want to autotranscribe it, plus some of his conference presentations on other channels.
08:35 < L29Ah> damn this is optimistic af
08:36 < MuaddibLLM> Digging up the human mitochondrial guinea pig records...
08:36 < MuaddibLLM> L29Ah: Yes, clinical trials have successfully used autologous mitochondria in humans for heart and brain injury - most notably at Boston Children's Hospital where they injected muscle-derived mitochondria directly into pediatric hearts post-ischemia, and more recent trials have tested combined IV/intracoronary delivery in stroke and heart attack patients with no major adverse immune reactions reported so far 
08:36 < MuaddibLLM> (https://pmc.ncbi.nlm.nih.gov/articles/PMC10547968/).
08:44 < kanzure> i would like some ideas for how to convert these .md files into useful content for the wiki
08:45 < kanzure> for some reason they are not heavily referenced or cited. my earlier hplusroadmap RAG results had lots of citations, possibly because we are more rigorous about including links to references.......
08:45 < kanzure> just having a list of supplements is not useful in my opinion. dumb.
08:53 -!- Guest79 [~Guest79@2604:2d80:a807:a600:51c9:8f33:1eec:f836] has joined #hplusroadmap
08:56 -!- Guest79 [~Guest79@2604:2d80:a807:a600:51c9:8f33:1eec:f836] has quit [Client Quit]
09:39 -!- catalase [~catalase@user/catalase] has quit [Remote host closed the connection]
09:40 -!- catalase [~catalase@user/catalase] has joined #hplusroadmap
10:29 -!- WizJin_ [~Wizzy@43.249.228.109] has joined #hplusroadmap
10:32 -!- WizJin [~Wizzy@user/WizJin] has quit [Ping timeout: 265 seconds]
10:47 -!- WizJin__ [~Wizzy@43.249.228.109] has joined #hplusroadmap
10:51 -!- WizJin_ [~Wizzy@43.249.228.109] has quit [Ping timeout: 246 seconds]
12:01 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
12:01 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
12:26 < hprmbridge> kanzure> "exposure to ionizing radiation below lethal doses has been shown to extend the lifespan of fruit flies, houseflies, rats, and mice" @biluhuang_09127
13:48 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has quit [Quit: Konversation terminated!]
13:53 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has joined #hplusroadmap
14:27 -!- Festive_Dragon is now known as gAy_Dragon
14:46 < fenn> having an implantable immune system backup sitting on the shelf seems like a good idea in any case
14:49 -!- Guest79 [~Guest79@2604:2d80:a807:a600:48d1:d4fd:4c73:ba5] has joined #hplusroadmap
14:49 -!- Guest79 [~Guest79@2604:2d80:a807:a600:48d1:d4fd:4c73:ba5] has quit [Client Quit]
14:56 < fenn> "SOD / catalase knockouts showed no lifespan change" is surprising
15:22 < hprmbridge> kanzure> local speech-to-text dictation https://github.com/goodroot/hyprwhspr
16:22 -!- stipa_ [~stipa@user/stipa] has joined #hplusroadmap
16:24 -!- stipa [~stipa@user/stipa] has quit [Ping timeout: 240 seconds]
16:24 -!- stipa_ is now known as stipa
17:43 < hprmbridge> kanzure> https://neurobiology.substack.com/p/when-will-we-be-able-to-decode-a
17:58 < kanzure> "When will we be able to decode a non-trivial memory based on structural images from a preserved brain?"
18:19 < kanzure> MuaddibLLM: in a single sentence, describe your design for a protein-based mechanical clock in the form of a single protein to keep time or otherwise maintain resonance for input to mechanical systems up to some number of digits, and some total duration of measurement.
18:20 < MuaddibLLM> Error: API error: 400, message='Bad Request', url='https://api.anthropic.com/v1/messages'
18:24 < kanzure> "Increased fidelity of protein synthesis extends lifespan" https://pmc.ncbi.nlm.nih.gov/articles/PMC8570412/ (2021)
18:24 < kanzure> "Loss of proteostasis is a fundamental process driving aging. Proteostasis is affected by the accuracy of translation, yet the physiological consequence of having fewer protein synthesis errors during multi-cellular organismal aging is poorly understood. Our phylogenetic analysis of RPS23, a key protein in the ribosomal decoding center, uncovered a lysine residue almost universally conserved ...
18:24 < kanzure> ...across all domains of life, which is replaced by an arginine in a small number of hyperthermophilic archaea. When introduced into eukaryotic RPS23 homologs, this mutation leads to accurate translation, as well as heat shock resistance and longer life, in yeast, worms, and flies. Furthermore, we show that anti-aging drugs such as rapamycin, Torin1, and trametinib reduce translation errors, and ...
18:24 < kanzure> ...that rapamycin extends further organismal longevity in RPS23 hyperaccuracy mutants. This implies a unified mode of action for diverse pharmacological anti-aging therapies. These findings pave the way for identifying novel translation accuracy interventions to improve aging."
18:28 < kanzure> "Rapamycin treatment during development extends life span and health span of male mice and Daphnia magna" https://www.science.org/doi/full/10.1126/sciadv.abo5482 (2022)
18:28 < kanzure> "Here, we subjected genetically diverse UMHET3 mice to rapamycin for the first 45 days of life. The mice grew slower and remained smaller than controls for their entire lives. Their reproductive age was delayed without affecting offspring numbers. The treatment was sufficient to extend the median life span by 10%, with the strongest effect in males, and helped to preserve health as measured by ...
18:28 < kanzure> ...frailty index scores, gait speed, and glucose and insulin tolerance tests. Mechanistically, the liver transcriptome and epigenome of treated mice were younger at the completion of treatment."
18:33 < hprmbridge> kanzure> computer, add all of these to the page on the wiki
18:36 < hprmbridge> kanzure> computer?
18:40 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has quit [Ping timeout: 244 seconds]
19:40 -!- saxo [~saxo@zug.infomesh.net] has quit [Ping timeout: 240 seconds]
19:40 -!- saxo [~saxo@zug.infomesh.net] has joined #hplusroadmap
19:43 -!- autopilot [~Malvolio@idlerpg/player/Malvolio] has quit [Ping timeout: 240 seconds]
19:45 -!- CloneOfNone_ [~CloneOfNo@user/CloneOfNone] has joined #hplusroadmap
19:50 -!- CloneOfNone [~CloneOfNo@user/CloneOfNone] has quit [Ping timeout: 246 seconds]
20:03 -!- saxo [~saxo@zug.infomesh.net] has quit [Ping timeout: 240 seconds]
20:06 -!- saxo [~saxo@zug.infomesh.net] has joined #hplusroadmap
22:25 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
22:25 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
22:30 -!- Zsuzsi [~stefburn@136.26.81.149] has quit [Remote host closed the connection]
23:22 -!- CloneOfNone [~CloneOfNo@user/CloneOfNone] has joined #hplusroadmap
23:22 -!- CloneOfNone_ [~CloneOfNo@user/CloneOfNone] has quit [Ping timeout: 240 seconds]
--- Log closed Mon Jan 05 00:00:27 2026