matteomics.bsky.social
Measurements want to be accurate;
Experiments want to be elegant;
Data wants to be beautiful and Data wants to be free
#proteomics
#rstats
81 posts
117 followers
386 following
Regular Contributor
Active Commenter
comment in response to
post
Signed up for the broadcast. It's a nice feature. I'm greatful it's available
comment in response to
post
Anyway, AI just summarizes what everyone else has written about their thoughts about the thing the AI itself has never, can never, see or experience. There are uses for that capability, but not original research.
comment in response to
post
There’s this quote I think about a lot:
“Research is to see what everybody has seen and think what nobody has thought.”
Attribution hard to pin down. Possibly Arthur Schopenhauer rephrased by Albert Szent-Györgyi.
comment in response to
post
More likely LLMs will be fed the politicians preferred decision and required to construct plausible sounding, and politically appealing, supporting arguments, based on selected historical precedents.
comment in response to
post
and govt policy promiting survival of the strongest will also be good, and improve the nation, and incidently continue to increase their wealth.
comment in response to
post
It feels to me that the billionare class has a deep need to feel their wealth is moraly justified and the soft eugenics policies are an effort to reify their internal justifications as policy. ie, their wealth is good because it examplifies the natural principle of the survival of the strongest,
comment in response to
post
I know right. Biologists tryng to remember physical chemistry!
comment in response to
post
You know what, it says why it's a F orbital in the blurb in the link.
comment in response to
post
So I looked at the cover art and thought, oh, an F shell. Then I thought, no Deep Six, must be a D shell, I'm miss remembering. So I looked it up and it is an F shell afterall. Is there a reason?
It's a very cool bit of science-art and I am very proud to have remembered my elecron orbital shapes.
comment in response to
post
Cool. I thought it ws probably AI generated. Is it OK if I steal it?
comment in response to
post
This is amazing, where is it from?
comment in response to
post
That is a point! It should say "The proverb is new to me but might mean..."
comment in response to
post
I've had a bit of fun with this today as well.
Alternative perspective: AI is providing plausible meanings to novel proverbs. Similar to the language games Douglas Adams used to inventing proverbs for Hitchhiker's. It's more sophisticated verbal reasoning than looking up the meaning of real ones.
comment in response to
post
Intersting that D₂O is lethal to mice in high relative proportion to H₂O, and probably most organisms. Obviosly lots of chemestry going on but I wonder if partial inactivation of most proteins, and total deactivions of some, plays a part in cause of death.
journals.physiology.org/doi/abs/10.1...
comment in response to
post
Want your son to make the sports team? How about some neanderthal muscle mass?
comment in response to
post
If you wanted to develop and showcase genetic modification for trait aquisition in mammals; with a view to offering to enhance human babies in the long run; inserting traits into modern animals so they resemble extinct charismatic mega fauna is the sort of thing you might do.
comment in response to
post
I feel quit deeply everyone just needs to get a moral compass. What are people doing faking the presence/absence of a spesific protein in a spesific mixture? Is that any way to spend a life?
comment in response to
post
OMG. I just love the bits from inside a mass spec. One of my colleagues has an orbitrap mass analyser on a shelf in their office of which I am very jealous.
comment in response to
post
I do agree really. The trick is to make all the work build intentionally towards publication. It's just the skill and effort required to do the work is pretty massive.
comment in response to
post
Ha. I know it's not. Just from the outside looking in I'm heart sick sorry for you all. I hope you can keep doing your science.
comment in response to
post
Oh, life will survive. It's civilisations with economies sophisticated enough to support stock markets that are in real danger.
Won't anybody think of the stock markets?
comment in response to
post
I fear the present moment in America is the culmination of a generational struggle by the far right. From the tea party, through to the Heritage Foundation and QAnon. Everything the liberal consensus so patiently built is in peril from an enemy thought beaten long ago.
comment in response to
post
I looked up the melting temperature of gold yesterday*. Google AI claimed gold melted at one of the highest temperatures for a metal. Wikipedia article below confirmed it's one of the lowest.
*There were convictions of thieves who stole a gold toilet - piqued my interest.
comment in response to
post
Hmm. The philosophy is out of my league. But I feel being human is a lot of wanting: being hungry and horney and cold and tired and lonely. I don't know if those things are calculations, or if AGI would need them. But without wants how could intelligence direct itself?
comment in response to
post
Hits nail on head.
Think it's an example of the general problem of falacy of believing incremental improvement in one tech will add up to a qualitative change. Like assuming we can go:
Write brothers
Biplane
Jet fighter
Space shuttle
Starship Enterprise
The final steps may be insurmountable.
comment in response to
post
So the thing Canada has that the US most plausibly needs is.... habitability when climate change really starts to bite. Combined climate change denial and climate change motivated land grab seems quite on brand. Maybe Trump wants to buy the future Newfoundland Riviera up cheap.
comment in response to
post
Adrian has a unique voice and if it did not exist I can't imagine anyone trying to simulate it.
comment in response to
post
There are already more books written by real humans than I could read in a lifetime. Why would I want to read the output of an algorithm trained on those books instead? Failure to understand the value of the art is the intentionality of the engaged human being producing it.
comment in response to
post
I love this idea! Was my thought when I found out the only evidence for the meg's size was the teeth... Maybe a goofy little fish with out of proportion teeth?
comment in response to
post
If they'd just put the hair gene into a kangaroo instead of a mouse at least they'd be half way to a Christmas cracker joke.
comment in response to
post
The cost saving implications of no samples per condition are a game changer.
comment in response to
post
That is a good point. Personally never occured to me that NCBI and refseq could ever actually be at risk. Possibly the EBI should look at complete replication for the sake of redundancy.
Sad US researchers wanting to replicate work in future need to consider risk of US databases disappearing.
comment in response to
post
So EMBL and www.ebi.ac.uk/uniprot/index are still there! Maybe time to sample the European tools and data sets! Personally I find the uniprot website richer and better designed than NCBI.
comment in response to
post
I usually re-heat it in the microwave, and forget to take it out again.