@modrak_m@bayes.club avatar

modrak_m

@modrak_m@bayes.club

Bioinformatics, biostatistics. Bayesian. Stan language, simulation-based calibration.
Currently at Deparment of Bioinformatics, Second Faculty of Medicine, Charles University in Prague

This profile is from a federated server and may be incomplete. Browse more on the original instance.

jonny, to random
@jonny@neuromatch.social avatar

What other disciplines have an extremely strong set of visual icons and imagery like neuroscience? I assume "all," but eg. In neuro we have

the sagittal human brain shape,
the 2d blobby neuron (usually pyramidal-ish with projections ~1-2x body diameter),
graphs/networks (ok this is lots of things but ours usually have visible/larger than average nodes)
For icons, for visual style we have the cajal sepia/black monochrome... etc.

Definitely my bias but its easier for me to imagine a neuron icon than like an earthquake or metaphysics icon. I also wonder whose iconography would be the most illegible to ppl outside it (and why is it going to be number theory)?

modrak_m,
@modrak_m@bayes.club avatar

@jonny Genomics: double helix, ATCG letters, mendelian diagrams; Molecular biology: the pipette, lambda phage, labcoat + goggles (that's a bit more generic though); Law: the paragraph symbol

jonny, (edited ) to random
@jonny@neuromatch.social avatar

Genuinely wondering how this is a crime. Ethereum is not a currency. This is like stealing 25 million robux or runescape coins
https://www.justice.gov/opa/pr/two-brothers-arrested-attacking-ethereum-blockchain-and-stealing-25m-cryptocurrency

Edit: question p much answered by @modrak_m in this post and replies:
https://bayes.club/

modrak_m,
@modrak_m@bayes.club avatar

@jonny @blake I am pretty sure that in most jurisdictions the most relevant consideration is the damage caused. So stealing robux would be a crime to the extent you convince the judge that it has value (not necessarily just a monetary one). For ethereum this is quite easy - many people are willing to exchange eth for money, so it almost by definition has value. Discovering a bug and responsibly disclosing is quite safe - there's no damage*

modrak_m,
@modrak_m@bayes.club avatar

@jonny *) there's some weird specific anti-hacking law in some countries (notably US), which makes this more complex, but my understanding is that this still allows for whitehat disclosure.

modrak_m,
@modrak_m@bayes.club avatar

@jonny **) Some attackers also try to claim that their actions couldn't be illegal because the software allowed them (in a sort of "code is law" argument). While there are some edge cases, my impression is that courts generally don't take those arguments well and will distinguish between what code does and what is its intended purpose and examine whether the attacker sincerely believed their actions were allowed and harmless. A recent case is e.g. https://www.web3isgoinggreat.com/?id=avi-eisenberg-convicted

modrak_m, to random
@modrak_m@bayes.club avatar

BMJ with IMHO sensible and simple take on multiple-testing corrections:
"Multiple testing should be adjusted for only where authors use the significance of statistical tests to weight the reporting, discussion, and interpretation of their findings."
There's a lot of nuance to this, but as far as simple summaries go, this is quite good!
https://doi.org/10.1136/bmjmed-2023-000783

Also in line with the recent paper by @MarkRubin on when not to adjust for multiple testing (https://doi.org/10.1016/j.metip.2024.100140).

jonny, to random
@jonny@neuromatch.social avatar

@ai6yr got me wondering what it would take to keep bees in the backyard of my LA apartment, and everything is telling me that they would probably be unhappy here bc very few flowers but i want a rack of little buzzers to go and hang out with :'( . one day i will move back to a place where the bees can be happy

modrak_m,
@modrak_m@bayes.club avatar

@jonny @ai6yr Not sure about LA specifically, but I've heard on several ocasions that bees can in fact be quite successful in inner cities. There definitely are some beehives on Prague rooftops, so....

modrak_m, to ai
@modrak_m@bayes.club avatar

Brutal, but IMHO quite likely correct:
"Sam Altman desperately needs you to believe that generative AI will be essential, inevitable and intractable, because if you don't, you'll suddenly realize that trillions of dollars of market capitalization and revenue are being blown on something remarkably mediocre." 1/2
https://www.wheresyoured.at/peakai/

modrak_m, to bioinformatics
@modrak_m@bayes.club avatar

A nice little paper arguing that Chord diagrams are generally harder to understand than Sankney plots for the same data. Relevant for when people will next request the "nice circular diagrams" for a bioinformatics analysis: https://dl.acm.org/doi/fullHtml/10.1145/3544548.3581119
#hci #plots #bioinformatics

A Sankney plot showing the flow of loans and debts between a few nations.

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • megavids
  • khanakhh
  • mdbf
  • ethstaker
  • magazineikmin
  • GTA5RPClips
  • rosin
  • thenastyranch
  • Youngstown
  • InstantRegret
  • slotface
  • osvaldo12
  • kavyap
  • DreamBathrooms
  • JUstTest
  • Durango
  • everett
  • cisconetworking
  • normalnudes
  • tester
  • ngwrru68w68
  • cubers
  • modclub
  • tacticalgear
  • provamag3
  • Leos
  • anitta
  • lostlight
  • All magazines
    •