"the bigger and more practically-usable hint of R's [Lisp] heritage is a little anomaly that every function returns whatever the last line says." — Ben Klemens, Math outside of time: A short lecture on math and computing. codeberg.org/b_k/math_out... [PDF]
At last I have got around to responding to Professor Dame Uta Frith's TES piece, in which she questions the autism spectrum
In a nutshell, I disagree...
"yes, autism is a broad and diverse category — that doesn’t mean there’s something wrong with the category itself"
medium.com/@suefletcher...
Macmind at work
A complete transformer neural network implemented entirely in HyperTalk, trained on a Macintosh SE/30.
github.com/SeanFDZ/macm...
When the asteroid hit, why did birds survive when all other dinosaurs died?
It's a question I ponder in my upcoming book The Story of Birds. But first, I explore it here for @sciam.bsky.social !
www.scientificamerican.com/article/an-a...
General, orders-of-magnitude faster whole-genome analysis with genotype representation graphs www.biorxiv.org/content/10.64898/2026.04...
"These days, if I suggest that you answer a question by consulting a computer, you would think I meant to go ask Google. Nevertheless, programming for answers is still a living art."
www.americanscientist.org/article/calc...
A generous donation from The Karlsson Játiva Charitable Foundation will fund 3 5yr fellowships in Brain Health and Multimorbidity at the Institute for Neuroscience & Cardiovascular Research:
2 x KJCF Quest Research Fellows and 1 x KJCF Quest CLINICAL Research Fellow.
edin.ac/4bZDPE9
Then what is all this SNP h2 we claim to estimate from data? I think the biggest problem in the literature is that the ideal quantity in which we try learn is vaguely specified. Is it an average? or is it the actual population's realized value? 6/n
We are now recruiting for new philanthropically funded fellowships in Brain Health and Multimorbidity at the University of Edinburgh, in INCR and @edinunineuro.bsky.social 🎉
There are two non-clinical posts and one clinical elxw.fa.em3.oraclecloud.com/hcmUI/Candid...
Please repost!
This paper reminded me that during my PhD I had made a hierarchical Dirichlet model of population differentiation of serotonin and dopamine receptor genes in long-tailed macaques.
New Perspective form Rory Maizels & me: "Gene regulatory networks: from correlative models to causal explanations"
Gene regulatory networks are supposed to give us mechanistic explanations of development, so why are we drowning in 'hairballs' of statistical correlations?
rdcu.be/e7zx7
IGC is looking for two Postdoctoral Research Fellows in Molecular Epidemiology for the MRC-funded grant 'Blood-and saliva-based DNA methylation biomarkers of disease' which will explore epigenetic biomarkers of organ ageing, proteins & disease. More info & apply here by 20 March 👉 edin.ac/4aYx7Pt
Cool seeing GPU acceleration in scientific computing that isn't "AI".
We are looking to recruit 2 postdocs in molecular epidemiology (3y and 4y posts) to work on a methylation/proteomic ageing project as part of a recent MRC grant award. @uoe-igc.bsky.social @genscot.bsky.social @edinunilbc.bsky.social
elxw.fa.em3.oraclecloud.com/hcmUI/Candid...
Relevant corollary: if you don't know all the nuances of the synonym, just tell me the word (prompt) you originally came up with.
I use LLMs a bit like a thesaurus.
The thesaurus can guide you to a synonym but you need to know what all the words mean to select the right one.
Author @suereviews.bsky.social recently joined @coolcatteacher.bsky.social on Cool Cat Teacher Talk to discuss the neurodiversity paradigm, the double empathy problem, and practical classroom strategies from her new co-authored book 'It Takes All Kinds of Minds.'
Listen here: spr.ly/6003hl7ZN
Two post-doc post in ARG space and loads of internal and external collaboration
Postdoctoral Research Associate in quantitative genomics
elxw.fa.em3.oraclecloud.com/hcmUI/Candid...
Postdoctoral Research Associate in quantitative genetics and breeding
elxw.fa.em3.oraclecloud.com/hcmUI/Candid...
Our new preprint “Learning lifetime disease liability reveals and removes genetic confounding in electronic health records” is now online! Link to paper: This work is led by my postdoc Yazheng Di and it’s our first project at @bsse.ethz.ch :) medrxiv.org/cgi/content/...
Thread 1/n
My Hopkins History dept. colleague François Furstenburg is JHU's gadfly— poking at hypocrisy and profiteering. But his biting critique applies broadly across the Academy.
Rx reading to anyone who cares about higher ed.
h/t @ayahnerd.bsky.social
i love the kinds of illusions that completely disappear as soon as you squint your eyes. they're like a tiny little light-hearted prank rather than complete destruction of everything you thought was real.
this is a very cool study, but i have to confess I'm reposting it because I want you to see the illustrations of these no-thoughts-in-head chicken participants
Genetics Society Anniversary Day(s). Lectures from honorary members and prize winners. 25 & 26 June, 2026 in Edinburgh.
genetics.org.uk/events/anniv...
Looks like a great line-up!
Schematic of a neuron showing sushi-belt model compartments. Neuron has two directional protein trafficing lanes shuttling proteins to four synapses.
"Sushi belt" model of synaptic protein diversity.
Protein diversity is spatially organised along dendritic trees of neurons, as explored and explained by this computational model.
www.biorxiv.org/content/10.1...
Wow! Happy I could help! But I'm sure I learned a lot more from you two than you could have from me. Super important work explaining who does the dirty work that makes up the logic of the authoritarian state.
#ethics #autocracy
Human brain development is not like mouse! Yet another exciting addition to our growing list of human cerebellar developmental programs different from mouse-centric dogma. We can't model what we don't know! Congrats @haldipur.bsky.social and our team - including many wonderful UW undergrads
