AI-for-Science Fellow at @BroadInstitute || Forbes 30u30 || Harvard Biophysics PhD '23 || ND Physics ‘17 || 🇺🇸🇭🇺🇸🇪🇷🇴 || My opinions, not my employer's
Boston, MA
- Tweets 1,878
- Following 294
- Followers 2,019
- Likes 3,114
Pinned Tweet
I love science, but I've found that a focus on productivity has softened the general curiosity I started grad school with. To remedy this, my 2023 Resolution is to read a paper a day, focusing on breadth. I'll be tracking my progress and thoughts below 👇
6
7
120
31,600
Recent discoveries on the acquisition of the highest levels of human performance by Güllich et al.
The best careers may not look optimal early.
World-class performers often explored more, and specialized later.
science.org/doi/10.1126/scie…
1
4
911
Cooperation by non-kin during birth underpins sperm whale social complexity by Maalouf et al.
Sperm whale birth is not solitary.
It is coordinated, communal, and cross-kin: a social network forming around a newborn calf.
science.org/doi/10.1126/scie…
1
6
881
Wild orangutans maintain sleep homeostasis through daytime naps by Ashbury et al.
Sleep may be “great nature’s second course,” but sometimes nature serves leftovers at lunch.
Wild orangutans compensate for short nights with daytime naps.
cell.com/current-biology/ful…
1
334
Microglia replacement halts the progression of microgliopathy in mice and humans by Wu et al.
Cell therapy for the brain, but not replacing neurons.
Replace the mutant microglia, slowing disease progressing.
science.org/doi/10.1126/scie…
2
16
61
4,053
See also the Perspective:
Replacing microglia to treat a brain disease by Du and Kipnis
A nice framing of why the brain’s immune cells may become a therapeutic target in their own right.
science.org/doi/10.1126/scie…
1
230
Senescence-resistant human mesenchymal progenitor cells counter aging in primates by Lei et al
Stem cell therapy for aging may not be about replacing tissues cell by cell.
Engineered progenitor cells seem to rejuvenate the aging environment in macaques.
cell.com/cell/fulltext/S0092…
1
1
4
707
Compartmentalization reduces conflict in multipartner plant-insect symbioses @Chomicki_G et al.
Mutualism as conflict managed by structure.
Fijian plants keep rival symbionts cooperating by building walls between them.
science.org/doi/10.1126/scie…
259
Hungary’s chance to rebuild science by @barabasi
Hungarian science has punched above its weight.
Now the harder task: building institutions worthy of that talent, protected from the politics that pushed so much of it away.
science.org/doi/10.1126/scie…
1
1
2
450
Hypothalamic clock governs circadian pain by Wei, Lou, Li et al.
Pain runs like clockwork.
A brain clock circuit rhythmically tunes how neuropathic pain feels.
science.org/doi/10.1126/scie…
2
1
21
1,196
See also related Perspective by Beggs and @Franklandlab:
Pain across time
science.org/doi/10.1126/scie…
1
185
Higher carbon storage in primary than secondary boreal forests in Sweden by Pascual et al.
(🇸🇪 pride bias)
Primary forests store ~72% more carbon than managed secondary forests.
science.org/doi/10.1126/scie…
2
2
283
NREM sleep improves behavioral performance by desynchronizing cortical circuits by Kharas et al.
Sleep has always felt like the ultimate biological tax.
Electrical stimulation mimicks sleep’s cognitive boost by shifting cortical population dynamics.
science.org/doi/10.1126/scie…
1
7
36
3,344
Language-specific neural dynamics extend syntax into the time domain by Coopmans et al.
Hungarian grammar feels like math in disguise.
Maybe that explains Erdős and Neumann: every sentence is a tiny structured inference problem.
journals.plos.org/plosbiolog…
14
42
2,864
Mastering diverse control tasks through world models by @danijarh, Pasukonis, Ba, and Lillicrap
A step toward agents that learn by imagination: build a model of the world, simulate possible futures, and train behavior inside them.
nature.com/articles/s41586-0…
5
30
2,441
Revealing how fungi build planet-altering “road” networks by Neo D. Martinez and Gianna L. Marschmann
Design principles for living supply-chain networks.
nature.com/articles/d41586-0…
1
5
489
Paper: A travelling-wave strategy for plant–fungal trade by Loreto Oyarte Galvez, Corentin Bisot et al.
Mycorrhizal fungi build trade networks as travelling waves: growing tips pull the frontier forward, while fusion regulates density behind it.
nature.com/articles/s41586-0…
218
Matching sounds to shapes: Evidence of the bouba-kiki effect in naïve baby chicks by Loconsole, Benavides-Varela, @LuciaRegolin
Baby chicks know that “kiki” is spiky and “bouba” is round.
Tiny, fluffy hint that sound-shape mappings may predate language.
science.org/doi/10.1126/scie…
1
3
310
As Perlman and @BodoWinter note in the accompanying Perspective, this begins to unlock how nervous systems connect structure across the senses.
Can't get enough of the paper's figures.
science.org/doi/10.1126/scie…
1
184
Whole-body molecular and cellular mapping of the laboratory mouse
Biology is usually measured organ by organ.
Clevenger et al make the mouse measurable as a whole-body molecular system.
Just don’t call it mouse carpaccio.
cell.com/cell/fulltext/S0092…
1
15
79
6,582
One of the strongest critiques of AI Scientists I've heard from Profs is that training on papers does not capture the 99 other narratives that explored and either failed or weren't written up.
Curious about this approach as a way of capturing that elusive scientific "taste"
1/ For nearly 350 years, science has communicated itself through one object: the paper. A linear narrative, frozen as a PDF, written for a human reader. We've come to treat that format as the medium of science itself.
It doesn't have to be. It's a historical artifact. 🧵
4
3
19
2,049