YouTubeでの反応が見たいという要望があったので代理投稿してもらいました
youtu.be/2Bro9nJFpks?si=KJXM…
第二ファシンテルン下北沢支部▐┛ retweeted
Old incels viewed themselves as outsiders whose failure to live up to societal standards gated them from a normal life. Nu incels view themselves as representing the common man who has been screwed over by collective female behavior.
第二ファシンテルン下北沢支部▐┛ retweeted
Chad was a hated demonic figure in Early Incel, but developed into a neutral deity in Middle Incel and entered the pantheon of Late Incel
草
>be comp biologist in 2005
>chimp genome just dropped
>everyone comparing genes, looking for protein differences
>nah, i have a better idea
>find DNA that mouse, rat, and chimp all agree on
>stuff that hasn't changed in 300 million years
>*ultraconserved*
>then check if it suddenly breaks in humans anyway
>run the scan
>49 regions light up
>only 2 of them code for proteins
>mfw the most human parts of the genome aren't even protein genes
>top hit: 118 bases
>chicken vs chimp: 2 changes in 310 million years
>chimp vs human: 18 changes in ~6 million
>name it HAR1
>human accelerated region 1
>cyberpunk name af
>"what does it do?"
>"we'll find out later"
>be wet lab biologist down the hall
>handed a 118-base mystery
>colleague in Brussels is staining embryonic brain tissue
>mail him a probe, "try this"
>months pass
>email arrives
>it lights up in Cajal-Retzius neurons
>7 to 9 weeks gestation, right when the cortex is building its layers
>it's an RNA gene, folds into a new shape in humans
>10 of the 18 changes keep the fold intact
>evolution was being careful
>still can't say what it actually does
>be enhancer lab in 2008
>run your own human accelerated scan
>top hit: HACNS1, another HAR
>stick human version into a mouse embryo with a blue dye reporter
>blue in the paw
>specifically where the thumb and wrist will form
>chimp version barely does it
>swap 13 human mutations into the chimp version
>now it does it
>we may have found part of the genetic story of the human hand
>still not sure of the general human accelerated function though
>be population geneticist
>see everyone celebrating "positive selection"
>notice the mutations are suspiciously biased toward G and C
>gc_biased_gene_conversion.jpg
>recombination can fake a selection signal for free
>build a model to separate them
>somehow the HARs survive the cut
>about a fifth of them are probably just gene conversion
>be the guy who brings this up at every conference
>nobody likes this guy
>but this guy is right
>be autism geneticist in 2016
>if these regions made the human brain
>breaking them should break something human
>check families where parents are related
>find HAR mutations disrupting both copies
>up to 5% of autism cases in those families
>some reviewers: "most of those CNVs hit genes too"
>shrug
>be machine learning guy, 2023
>train a model to read enhancers
>score every human-chimp difference in ~2,600 HARs
>expect all mutations to push the same direction
>43% of HARs have mutations fighting each other
>one turns the enhancer up, the next turns it down
>evolution overshot and corrected
>let's keep pushing
>huh, human DNA folds differently in the nucleus than chimp DNA
>structural rearrangements over a million years ago
>suddenly old enhancers are next to new genes
>"enhancer hijacking"
>post it, feels good, finally solved
>be yale lab, 2025
>map what 1,590 HARs actually touch, human and chimp side by side
>nearly 3,000 targets
>almost all shared between species
>no hijacking, just retuning old genes
>say the other guys were wrong
>post it, this is *not*solved
>be 2025
>we have CRISPR, organoids, and humanized mice
>put human HARs in chimp neurons
>they grow more branches
>put human HARE5 in mice
>more neurons, bigger cortex
>delete HAR123
>mice get worse at unlearning things
>finally testing them one at a time
>~3,000 to go
>be present day, present time
>telomere-to-telomere genomes for every great ape
>find whole new rapidly evolved regions that weren't even in the old reference
>call them HAQERs
>mfw more HARs
>the list keeps growing
>these things keep popping in every genome scan
>mfw the only thing accelerating here is our confusion
>mfw we still don't fully know what HAR1 does
3
996
投稿しました
Japan, Communist & Party - Moccomintern
nicovideo.gay/watch/sm469960…
#sm46996011
第二ファシンテルン下北沢支部▐┛ retweeted
INMU KINGにだまされる人は、芸術の教養が根本的に欠けている。クラシックからロック、ポピュラーまで、音楽のほんものに触れていれば、AI楽曲には騙されない。INMU KINGを聴くんだったら、モーツァルトやビートルズ、ボブ・マーリーを聴いた方がはるかに深く世界に通じる教養が身につく。
モロッコレアで草
それを言いたかっただけだろ
[AI]Akira, Jeanne and Bouto - Moroccorea
nicovideo.jp/watch/sm4694895…
#sm46948952
#ニコニコ動画
1
654
おお
JUST IN: Palmer Luckey's genetically modified military dog startup, Warhound, comes out of stealth, aiming to create enhanced canines for the U.S. military.
Readers added context they thought people might want to know
This is NOT Palmer's company at all, his company Anduril is only supplying the tech for the dogs, and this is a wholly owed Collosal entiity
x.com/palmerluckey/s…
1
417























