--- Log opened Fri Mar 04 00:00:16 2022
00:22 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
00:39 -!- spaceangel [~spaceange@ip-78-102-216-202.net.upcbroadband.cz] has joined #hplusroadmap
02:25 < faceface> fltrz: I remember that story too, but I forget who it was
02:30 < faceface> Is anyone good at genome analysis? I'm trying to work literature on covid 'host genetics' into a persaonal genome report
02:31 < faceface> e.g. based on your 23andMe results, your 30% less likely to get a severe covid infection than someone 'like you' <- where that includes a bunch of phenotype data 
02:31 < faceface> It's not really hard, there is just a lot of literature to go through, and my code sux
02:32 < faceface> https://github.com/Geromics/covcheck/tree/feature/score/pmid/33888907 < first real attempt, but it's nasty
02:33 < faceface> three papers report SNPs in a haplotype that is in LD with a 23andMe v3 SNP, all three reports suggest a 50% decrease in risk (or 2x increase in risk depending on how the effect is defined)
02:34 < faceface> I'm 'cataloging' papers here: https://docs.google.com/spreadsheets/d/1ieO4jzxlyCtCzRKHywEWboD1s4zMdKJDtepmXVj6CC0/edit#gid=1926235615 
02:42 < nsh> <faceface> three papers report SNPs in a haplotype that is in LD with a 23andMe v3 SNP, all three reports suggest a 50% decrease in risk (or 2x increase in risk depending on how the effect is defined)
02:42 < nsh> there's your problem right there lol
02:43 < nsh> we must pursue a global armistice on the abuse of statistical modelling 
02:50 -!- juri_ [~juri@178.63.35.222] has joined #hplusroadmap
03:12 < faceface> nsh: what's the problem?
03:13 < nsh> ah i think i misread you actually
03:13 < faceface> no probs, I'm quite possibly doing it wrong!
03:13 < nsh> but if you can define an effect so it's either half as affected or twice as affected by some substitution then you probably don't have a great model and more generally i am doubtful of the entire methodoloy
03:13 < nsh> but it's not my circus
03:15 < faceface> The methodology is indeed 'dubious', but it's the current 'standard' in genetic analysis. Define two groups or some measure. Get SNP-Chip data from a few thousand participants. Do some stats. Report the markers.
03:17 < faceface> I don't think that the 'direction' of the effect is a big problem, it just depends on how you pick the allele. In general it will be something like, humans have an A at this position 20% of the time and a G 80% of the time in population x. If you carry an A you're less likely to have phenotype z in our study... It's equally valid to say, if you carry a G you're more likely to have phenotype z... at 
03:18 < faceface> least I think...
03:18  * nsh nods
03:18 < nsh> makes sense
03:21 < nsh> but has anyone actually thought about this with anything resembling a thorough information theoretic analysis? like even rudimentary coding theory makes it clear that you can't necessarily fish out any meaningful causal model by correlating binary distinctions on two sides of a system of enormous emergent complexity 
03:21 < nsh> if there were anything resembling comprehensive error correction then you already decouple the logical inputs from the coding inputs
03:35 < faceface> Yes indeed, it's complex 
03:35 < faceface> I'm just listening to a vid about the effect of gut microbiome, which is a big 'confounding factor' for one thing
03:35 -!- mirage335 [~mirage335@64.79.52.86] has quit [Quit: Client closed]
03:36 -!- mirage335 [~mirage335@64.79.52.86] has joined #hplusroadmap
03:36 < faceface> At one point I wanted to make a database of all metabolism related SNPs, and try to map metabolic outcomes onto that network... a bloot metabolite screen is 'only' about 100 USD per sample
03:36 < faceface> If you could sequence and analyse a few 100,000 people, it could be informative... perhaps...
03:37 < faceface> My thinking was that to 'drive' the system, you'd enroll people and then post them 'hyperdoses' of certain nutritional metabolits to pull out strong effects
03:38 < faceface> https://www.youtube.com/watch?v=58IhKxBQQpw <- talks about microbiome from about half way through
03:39 < faceface> However, regardless of the complexity, there are some clear associations between SNPs and phenotypes, e.g. BRCA2 and brest cancer, for example.
03:40 < faceface> Other studies have looked at propensity for infection... one moment
03:41 < faceface> https://www.nature.com/articles/nrg3114
03:44 < faceface> The big unknown in GWAS studies is usually weather the variation you find with the highest statistical assoication is actually causative or just 'tagging' the genetics in question. e.g. for gene therapies, changing the tag won't reduce your succeptibility, for example
03:45 < faceface> also there are weird effects where a SNP is linked to a trait in one 'background' such as in a european population, and that association just goes away in another genomic background.
03:45 < faceface> Which leaves you questioning, where is the actual biology in all this?
03:51 < faceface> This the main paper by the way: https://www.nature.com/articles/s41586-021-03767-x Figure 2 shows the stats I was talking about. The latest datset from the same study is presented here: https://www.medrxiv.org/content/10.1101/2021.11.08.21265944v2
03:52 < faceface> Basically I'm looking for people to help colate the data into a 'score' for anyone to take
03:57 -!- mirage335 [~mirage335@64.79.52.86] has quit [Quit: Client closed]
03:57 -!- mirage335 [~mirage335@64.79.52.86] has joined #hplusroadmap
04:04 -!- mirage335 [~mirage335@64.79.52.86] has quit [Quit: Client closed]
04:04 -!- mirage335 [~mirage335@64.79.52.86] has joined #hplusroadmap
04:07 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0::bb3b] has joined #hplusroadmap
05:27 -!- fltrz [~ik@109.236.138.239] has quit [Quit: leaving]
06:51 -!- nq_ [~nq@82.78.22.133] has joined #hplusroadmap
06:51 -!- test_ [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
06:54 -!- Hoolootwo [~Hooloovoo@hax0rbana.org] has joined #hplusroadmap
06:54 -!- archels_ [~neuralnet@static.65.156.69.159.clients.your-server.de] has joined #hplusroadmap
06:59 -!- Netsplit *.net <-> *.split quits: Hooloovoo, archels, _flood, catalase, nq
07:03 -!- catalase [catalase@stable.bnc.by.bnc4you.xyz] has joined #hplusroadmap
09:51 -!- mgxm [~mgxm@user/mgxm] has quit [Quit: ....]
09:51 -!- mgxm [~mgxm@user/mgxm] has joined #hplusroadmap
12:52 -!- mirage33561 [~mirage335@64.79.52.86] has joined #hplusroadmap
12:56 -!- mirage335 [~mirage335@64.79.52.86] has quit [Ping timeout: 256 seconds]
12:57 -!- mirage3356191 [~mirage335@64.79.52.86] has joined #hplusroadmap
13:01 -!- mirage33561 [~mirage335@64.79.52.86] has quit [Ping timeout: 256 seconds]
13:07 -!- mirage335619140 [~mirage335@64.79.52.86] has joined #hplusroadmap
13:12 -!- mirage3356191 [~mirage335@64.79.52.86] has quit [Ping timeout: 256 seconds]
13:16 -!- baracurda [~baracurda@user/parserblabla] has joined #hplusroadmap
13:39 -!- test_ is now known as _flood
14:01 -!- Codaraxis [~Codaraxis@user/codaraxis] has joined #hplusroadmap
14:17 -!- helleshin [~talinck@108-225-123-172.lightspeed.cntmoh.sbcglobal.net] has quit [Ping timeout: 245 seconds]
14:18 -!- helleshin [~talinck@108-225-123-172.lightspeed.cntmoh.sbcglobal.net] has joined #hplusroadmap
14:18 < lsneff> Been doing a bunch with amaranth recently
15:09 -!- spaceangel [~spaceange@ip-78-102-216-202.net.upcbroadband.cz] has quit [Remote host closed the connection]
15:17 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
15:19 -!- _flood [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 256 seconds]
16:05 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 240 seconds]
16:49 -!- mrdata [~mrdata@user/mrdata] has quit [Ping timeout: 240 seconds]
17:02 -!- mrdata [~mrdata@user/mrdata] has joined #hplusroadmap
17:29 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0::bb3b] has quit [Quit: Leaving]
20:34 -!- nq_ [~nq@82.78.22.133] has quit [Quit: Leaving]
21:07 -!- benthumb2000 [~paul@2400:4052:10a1:f200:f1e7:c618:7386:d87] has joined #hplusroadmap
23:02 -!- benthumb2000 [~paul@2400:4052:10a1:f200:f1e7:c618:7386:d87] has quit [Quit: WeeChat 3.3]
23:38 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
--- Log closed Sat Mar 05 00:00:16 2022