algobaker

A year here and he still dreamed of cyberspace, hope fading nightly. All the speed he took, all the turns he'd taken and the corners he cut in Night City, and he'd still see the matrix in his dreams, bright lattices of logic unfolding across that colourless void... The Sprawl was a long, strange way home now over the Pacific, and he was no Console Man, no cyberspace cowboy. Just another hustler, trying to make it through. But the dreams came on in the Japanese night like livewire voodoo, and he'd cry for it, cry in his sleep, and wake alone in the dark, curled in his capsule in some coffin hotel, hands clawed into the bedslab, temper foam bunched between his fingers, trying to reach the console that wasn't there.

@josiezayner That's a crazy reaction to this legitimately impressive announcement unless it's meant satirically

Don't we have enough of these biotech companies already?

@eryney_ok @ashleevance ...any new group designing a new payload will have to choose between paying the tax to use an existing capsid they see can already do the job, vs paying to do the experiments necessary to patent-bust it?

@eryney_ok @ashleevance Follow-up question though not sure if it's within things you've thought about. How will the economics of capsid designs end up working out? Will there be libraries of patented designs, and ... (1/n)

.@eryney_ok is brilliant, and he's here to answer your bio-tech and health questions. come one, come all corememory.com/p/ask-almost-a…

- David Goggins What if it actually works out?

We are all our own worst haters and doubters because self doubt is a natural reaction to any bold attempt to change your life for the better. You can’t stop it from blooming in your brain, but you can neutralize it, and all the other external chatter by asking, What if? (2/n)

The most important conversations you’ll ever have are the ones you’ll have with yourself. You wake up with them, you walk around with them, you go to bed with them, and eventually you act on them. Whether they be good or bad. (1/n)

@seaamiacsr "grinding 20 hours a day" we have very different biologies

introspecting within the tech abyss episode 1 - we say we want to make it. but there’s always that voice, cynical and annoyingly calm, telling us we’re not built for this life. that we’re built to read, to write, to live quietly.. by the shore, the walden kind of life. a life that would put an end to our capitalism-wired-anxiety and to our disguised greed. this life keeps showing up.. and it feels more honest (?). building a small, joyful family. teaching math in some unknown university. living lightly instead of grinding 20 hours a day trying to stand among the greats - the same greats you sometimes can’t help but despise. and yet, leaving like this atm feels so wrong. unfinished-half-used, not even half. like stopping before you even find out what you were actually capable of - oh actually before you find out anything at all. so you stay in the contradiction (or hypocrisy). you want out, but only after you make it. so the only optimal path is to do it anyway.. obsessively, crazily. because you have the capacity, and there’s no excuse not to try at least once in your life and avoid any late-regrets at 40. start now. or quit now. but stop being average. pick a side. for those on this side, build. write, study, work. be aggressive about whatever your definition of making it is. be loyal to yourself.. and make that a habit. for the ones on the shore, enjoy and avoid gazing into the abyss.

@LindorffLarsen Thanks for the paper - it's a great question to go after

@algobaker Thanks for the question. It certainly suggests that the disagreement with these MAVEs is perhaps not just because the VEPs are imprecise, though it was hard to test that explicitly.

New preprint on how disagreement among variant effect predictors (VEPs) can help guide prioritization of proteins for experimental analysis We analyse for which proteins VEPs disagree, what features they have, and suggest lack of concordance & clinical data to guide experiments

@robindchnt If you're staying in the city you are doing it wrong!

I try to run in every city I visit. Nothing comes close to Zurich. Not even close. Best running routes and air!

@effectfully There's more to life than money

bro

@anshulkundaje @MichaelLinLab @BoWang87 If the thing you're posting doesn't actually work, then real scientists aren't the users, and the post's intended audience must therefore actually be VCs and randos on X

@MichaelLinLab @BoWang87 Vast majority of the true users of these kinds of approaches (serious biologists & scientists) are not VCs, randos on X monetizing hype & gullibles who are going to fall for the hype. U end up getting 1000s of clicks & likes from all the wrong ppl, which makes u feel great.

Hype vs reality Hype: an advertising video with loud music. Hype: the word "reasoning".

@LindorffLarsen Then when that analysis shows there is no relationship, you seem to just say screw it "one should pick protein targets for MAVE based on disagreement between VEPs anyway". Doesn't the lack of correlation suggests this is no more informative than just picking random targets?

@LindorffLarsen Didn't get the logic of this one. You propose an interesting hypothesis "one should pick protein targets for MAVE based on disagreement between VEPs". Then you do the natural analysis, to see if the disagreement between VEPs correlates with errors in predicting MAVEs

@pranamanam @Ligandal lmao 'your work will be my point of reference for what not to do'

@Ligandal Btw, when I tell my students why they should do a PhD (to learn how to design, execute, write and communicate strong research with appropriate validations), I will point to this paper. 👆 Maybe the best advertisement for NOT skipping a PhD.

I usually don't like to criticize papers on social media, but this one deserves it. Not familiar with @Ligandal, but so many problems: AI-hallucinated citations, figures, no real validation, not "structure-free", and definitely not diffusion. I'll go thru my criticisms below. 👇

@iskander @Presidentlin It's a bit like the 'sent from my iPhone' thing the old iphones had built into the email signature

@Presidentlin Actually, how do I disable this? (It’s very weird and presumptuous)

Turns out people hate the Claude Code, adding its name to the group project.

@moleculesrcool @NousResearch Why does it say 'claude-opus-4.6'?

@NousResearch maybe this is old news but Hermes is effectively running bioreactors now

@jaiselsingh , I can only speak for the 10 years I actually spent in the field, where people said stuff like this all the time

@jaiselsingh Again, the statement we're discussing is about people overgeneralising to a whole cohort of researchers, so picking 3 guys in a particular era who didn't match the pattern doesn't really show anything. I have no idea if people said stuff like that back then (1/2)

there used to be a word for what drove the best researchers; not really ambition but obsession/ being haunted by their problems (i.e Demis) they couldn't stop working on the problem. now the field is full of people who picked their problem from a menu (related quote below)

@jaiselsingh I wasn't alive then so I can't confirm or deny. I'm started working 1 generation previous in machine learning research (post AlexNet, pre LLMs) and the current "picked their problem from a menu" comments were going on just as much then too

@algobaker i don’t agree with that tbh. nobody said this about Shannon or Feynman’s generation. some cohorts genuinely advance the field and some ride the wave. the ‘every generation’ framing only works if you don’t check