--- Log opened Tue Mar 25 00:00:55 2025
00:07 < fenn> adjusted for BMI is also very sus
00:10 < nsh> there's very little methodologically wrong with the study
00:10 < nsh> (afaik without reading the paper)
00:10 < nsh> (lol)
00:11 < nsh> it's just a longitudinal questionnaire. people might not report what their grocery store receipts would report but neither are they colluding somehow to make patterns appear
00:11 < nsh> the patterns are real
00:11 < nsh> take from it what you want
00:11 < nsh> the real issues are confounding factors as much if not more than the quality of the self-reporting
00:12 < nsh> (and choice of questions)
00:12 < nsh> if you find a product review useful information
00:12 < nsh> then this should be useful information with significant statistical quality
00:15 < jrayhawk> (if all of the product reviews were six months after their last interaction with the product, and also the products involved were some of the most fundamental parts of social identity)
00:30 < nsh> mmm
00:30 < fenn> why don't we use receipts instead of questionnaires?
00:31 < fenn> i know exactly how much of food X i eat because i don't throw any away
00:33 < jrayhawk> i have often wondered about starting an insurance company that offered discounts to those who showed receipts
00:34 < fenn> if it's a data regularization problem, we now have the tools (LLMs) to reason quantitatively about the contents of "a chicken salad"
00:34 < fenn> i mean they could have collected this info even if it were not usable at the time, with the hope that someone could go through and make sense of it later
00:40 < fenn> also frank hu https://www.theatlantic.com/magazine/archive/2023/05/ice-cream-bad-for-you-health-study/673487/
00:40 < fenn> .t
00:40 < saxo> Could Ice Cream Possibly Be Good for You? - The Atlantic
00:45 < jrayhawk> awkwardly, handling receipts is a liability in and of itself https://scholar.google.com/scholar?q=thermal+paper+handling+hormone
00:45 < fenn> right, that shit needs to get banned properly (not "BPA free" but "printed with ink")
00:46 < jrayhawk> that'd be nice
00:47 < fenn> has RFK done anything useful yet
00:47 < jrayhawk> not that i have seen
00:58 -!- flyback [~flyback@2601:540:c701:900:a72:bf40:85a5:dae1] has quit [Ping timeout: 260 seconds]
01:02 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has joined #hplusroadmap
01:11 -!- flyback [~flyback@2601:540:c701:900:8762:bd03:d9f3:3224] has joined #hplusroadmap
01:33 < fenn> hm. anecdata showing keto stops non-24 circadian rhythm disorder https://www.exfatloss.com/p/holy-s-and-t-is-my-non-24-gone
01:38 < fenn> uh except for a rice + marinara fat-free diet
01:40 < fenn> "it’s very nice to be able to have a social life and a career and all that stuff"
01:44 < fenn> "surgically activate my Non-24, cycle back around to normal work hours, and go back on keto to stabilize it." this is some real shit
02:03 < jrayhawk> big swings in choline and precursor availability to different cells with shifts between lipoproteins based on ApoA, ApoB48, and ApoB100 come to mind
02:04 < fenn> what's choline got to do with it
02:04 < jrayhawk> triple methyl donation to tetrahydrobiopterin
02:04 < fenn> also don't understand what lipoprotein has to do with it
02:07 < jrayhawk> has to do with which?
02:08 < fenn> non-24, presumably melatonin biosynthesis
02:08 < fenn> honestly my eyes bounce off Apo* like chinese characters
02:09 < jrayhawk> melatonin and also SCN confidence and precision via dopamine and acetylcholine
02:30 < jrayhawk> i say this while being awake four hours later than expected after not eating eggs for dinner
02:32 < hprmbridge> kanzure> god damn it, eat your eggs
02:37 < fenn> fortunately choline is cheap and readily available as a chemical dietary supplement
02:37 < fenn> i didn't know that BH4 was a cofactor in melatonin synthesis
02:38 < fenn> still waiting for a cyberspace interface to the known metabolome
02:38 < fenn> wikipedia rabbitholing is not ideal
02:39 < fenn> all the stupid names and words and symbols, just use chemical structure diagrams please
02:39 < hprmbridge> kanzure> what do you want your interface to do or show?
02:40 < fenn> the network of biochemical reactions that transform dietary inputs to bioactive molecules
02:41 < fenn> there's just so many facts that go into a metabolic pathway, it's hard to visualize
02:41 < fenn> i also want to be able to tweak a value and see the predicted effect
02:41 < hprmbridge> kanzure> https://reactome.org/PathwayBrowser/
02:41 < fenn> i realize this may be asking for too much
02:43 < fenn> "pathway browser" looks like an ontology browser, not a chemical reaction network browser, i.e. completely useless and stupid
02:45 < hprmbridge> kanzure> used to be something else
02:45 < fenn> there *are* some good diagrams out there, like KEGG
02:46 < fenn> bah the roche poster is offline
02:47 < fenn> "We're currently revising our Biochemical Pathways resource, a trusted reference for over 50 years. This revision will ensure it remains accurate and up-to-date"
02:48 < jrayhawk> if you do decide to screw around with that pathway, keep in mind maps won't show that DHFR is used in recycling dihydrobiopterin back up to tetrahydrobiopterin and is subject to competitive inhibition from folate cycling
02:50 < fenn> i have done self-experiments in the past with folate and methylfolate and didn't notice any effect on circadian rhythm
02:51 < fenn> admittedly it's hard to study because of the long time needed to make an observation
02:53 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Write error: Broken pipe]
02:53 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
02:54 < jrayhawk> and the inverse-u-shaped dose-response curve of simultaneous methyl donation from 5-MTHF and competitive biopterin inhibition from it bouncing around in the folate cycle, and a lot of individual genetic variability on all the enzymes involved
03:02 < fenn> ok the reactome pathway browser does contain reaction networks, i don't know why it didn't load the first time: https://reactome.org/PathwayBrowser/#/R-HSA-209776
03:02 < fenn> it's still very awkward and frustrating
03:06 < fenn> i wonder why thyroxine synthesis is on the same diagram, it looks to be completely disconnected
03:19 < fenn> these diagrams are still hand-made and don't include all known ways biochemicals affect subsystems. it just feels wrong. also there's no quantitative values like reaction rates or binding constant or arrhenius constant, typical rate limiting substrate or cofactor or enzyme. these should be communicated with things like sizes or line weights or proximity
03:20 < fenn> it really should feel more like swimming around in a web of goop with all the info right there, not clicking on a poster and looking at new tabs for basic info like chemical structures
03:40 < jrayhawk> a real-time XVIVO-style rendering system combined with vrui-style interactivity would sure be nice
03:41 < fenn> i just meant force directed graph layouts, but sure, nothing wrong with some fun graphics
03:43 < fenn> i'm partial to david goodsell's style
03:43 < kanzure> https://biocyc.org/MetNetExplorer.shtml?orgid=ECOLI
03:44 < fenn> these things really love opening new tabs unexpectedly
03:45 < kanzure> https://escher.github.io/#/app?map=RECON1.Amino%20acid%20metabolism%20(partial)&tool=Viewer
03:47 < kanzure> https://pathways.embl.de/ipath3.cgi?map=metabolic
03:49 < kanzure> this would be a cool hackathon theme.
03:49 < kanzure> get bryan johnson on that if he wants to do something useful
03:50 < nsh> lol
03:51 < kanzure> https://metabomaps.brenda-enzymes.org/maps.php?org=9&svg=72
03:52 < kanzure> https://metabolicatlas.org/explore/Human-GEM/map-viewer/golgi_apparatus
03:52 < fenn> the embl one shows chemical structures, although it's hidden behind a secondary tooltip for no good reason
03:55 < fenn> i really do think structures are critical for understanding wtf is going on
03:57 < fenn> imagine trying to figure out what's going on here without a structure diagram https://en.wikipedia.org/wiki/Lanosterol
03:57 < fenn> also 90% of the words in biology are just completely made up arbitrary bullshit
03:57 < fenn> we shouldn't respect that
04:16 < kanzure> https://js.cytoscape.org/
04:16 < kanzure> hm nevermind. that's graph theory focused.
04:17 < kanzure> pick your poison https://www.jointjs.com/demos
04:20 < fenn> i'd need to learn way too much hairy bioinformatics stuff to actually make this useful
04:23 < kanzure> i like the metabolicatlas.org or escher/embl style. 
04:23 < kanzure> compared to the others
07:00 -!- andytoshi [~apoelstra@user/andytoshi] has quit [Quit: Lost terminal]
08:22 < RangerMauve> cyberspace interface for the known metabolome is on my TODO list (probs years away). Currently messing with how to render and navigate graphs with hierarchies in 3D. Though I'm gonna be spending a bunch of time work on the UX of graph transformation too.
08:23 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has quit [Quit: Avoid fossil fuels and animal products. Have no/fewer children. Protest, elect sane politicians. Invest ecologically.]
08:24 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has joined #hplusroadmap
08:44 -!- potatope [sid139423@id-139423.lymington.irccloud.com] has quit [Ping timeout: 246 seconds]
08:47 -!- potatope [sid139423@id-139423.lymington.irccloud.com] has joined #hplusroadmap
10:23 < hprmbridge> Eli> How useful are these pathways do you think? 50% of researchers have experienced a situation where they couldn’t replicate their own experiment. And even stuff that’s hyper studied has a ton of mysteries we haven’t figured out. Then include the fact that genetics is absurdly messy so reactions are weird for everyone.
10:24 < hprmbridge> Eli> Some of our best selling medications have MOAs that are probably as accurate as saying “it happened because mars was in retrograde”
10:34 < hprmbridge> Eli> When I was at SXSW I saw a company that was selling a blood test that measured like 500 metabolites. I feel like systems biology could answer some of the interesting questions about being in a healthy state vs being in a diseased state.
11:06 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
11:06 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
11:12 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has quit [Quit: Konversation terminated!]
11:13 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has joined #hplusroadmap
11:19 < geneh2> https://www.technologyreview.com/2025/03/25/1113611/ethically-sourced-spare-human-bodies-could-revolutionize-medicine/
11:45 < kanzure> ethically sourced... how about people who sell their body parts? oh is making money not ethical?
11:50 < jrayhawk> Legalizing a free market for organs has too many pathological incentives. We don't even let doctors who know the organ donation status of patients do diagnosis and treatment due to how disastrous the incentives are.
11:54 -!- Gooberpatrol_66 [~Gooberpat@user/gooberpatrol66] has joined #hplusroadmap
11:54 < kanzure> do you say the same thing about legalized euthansia
11:55 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has quit [Ping timeout: 276 seconds]
12:00 < jrayhawk> Depends on what you mean. Fully liberalized suicide would mean a body with a gunshot wound would no longer qualify as reasonable suspicion of a crime having been committed for the purposes of the forth amendment.
12:02 < jrayhawk> er, fourth
12:03 < jrayhawk> this is, similarly, why you need a license to transport a dead body.
12:11 < kanzure> https://diyhpl.us/~bryan/projects/exitbid/
12:11 -!- Gooberpatrol_66 [~Gooberpat@user/gooberpatrol66] has quit [Quit: Konversation terminated!]
12:12 -!- Gooberpatrol_66 [~Gooberpat@user/gooberpatrol66] has joined #hplusroadmap
12:41 -!- andytoshi [~apoelstra@user/andytoshi] has joined #hplusroadmap
13:16 < hprmbridge> Eli> Also, I sort of like how these guys visualize pathways: https://iwasa.biochem.utah.edu/metabolism/chapter2.html
13:28 -!- andytoshi [~apoelstra@user/andytoshi] has quit [Ping timeout: 268 seconds]
13:46 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has joined #hplusroadmap
13:47 -!- Gooberpatrol_66 [~Gooberpat@user/gooberpatrol66] has quit [Ping timeout: 260 seconds]
14:03 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has quit [Quit: Avoid fossil fuels and animal products. Have no/fewer children. Protest, elect sane politicians. Invest ecologically.]
14:04 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has joined #hplusroadmap