iiinova — where your digital self keeps evolving

Reposted by Bruce Stagg Martin

Jhan C. Salazar, PhD @jhancsalazar.bsky.social · 06/10/2026

Happy to share that the last chapter of my dissertation is finally out in Evolution! 🦎⛰️ I'll share a longer thread soon about what we did and found. Huge thanks to all my coauthors: @jblosos.bsky.social, @juvelas.bsky.social, Steven Poe and Adam Algar doi.org/10.1093/evol...

doi.org

Repeated evolution of high-elevation occupation in tropical lizards

Abstract. Since Darwin and Humboldt, researchers have sought to understand how species adapt to elevation. Although tropical reptiles are typically most di

16330

Bruce Stagg Martin @bstaggmartin.bsky.social · 04/10/2026

Oh man I've been wanting to play around with webr for a while now--this certainly makes it all the more tempting!

100

Bruce Stagg Martin @bstaggmartin.bsky.social · 28/09/2026

Probably was just too long to fit in the margins

010

Reposted by Bruce Stagg Martin

Dr. Davey F. Wright ⛏️🦕🧬 @daveyfwright.bsky.social · 16/09/2026

"High-quality revisionary taxonomy must be grounded in experienced character evaluation and homology assessment, concept delimitation, and contingent analytical judgment, skills that are developed over years of immersive engagement with natural history collections" 👏👏👏

0215

Bruce Stagg Martin @bstaggmartin.bsky.social · 13/09/2026

Fucking finally someone said the quiet part out loud haha

000

Reposted by Bruce Stagg Martin

Cristian Román-Palacios @cromanpa.bsky.social · 09/09/2026

When two clades evolve together, can the trees alone tell you how often they cospeciated, and how often the symbiont switched hosts? We think so! Our paper, led by @yichaozeng.bsky.social, shows that branch length differences in the cophylogeny are informative enough to estimate six rates at once

03312

Bruce Stagg Martin @bstaggmartin.bsky.social · 09/09/2026

...the "rare exceptions" stop existing because they don't into a model. I thought we were just talking about you not being such a judgmental dick to people not fitting neatly in the binary.

010

Bruce Stagg Martin @bstaggmartin.bsky.social · 09/09/2026

"batshit" is correct! There's also the issue where you talk about DSDs and what not but people go "those are rare exceptions". I'm just like...yeah, what are we doing here? Are you trying to develop oversimplified models assuming a binary for drug dev or something? Go ahead! It's not like...

110

Bruce Stagg Martin @bstaggmartin.bsky.social · 09/09/2026

Annoying how physicists often seem to get a pass whenever they say "it's more complicated than your common sense would suggest" while us biologists are told we're crazy for possibly suggesting their "common sense" regarding sex might just be oversimplified 🤦

020

Bruce Stagg Martin @bstaggmartin.bsky.social · 09/09/2026

...case-by-case basis anyways (well, at least to the exent that such decisions can be made without bias, which seems dubious in the current cultural climate). I could go on about how much this stuff drives me crazy, but I'll leave it at that for now.

000

Bruce Stagg Martin @bstaggmartin.bsky.social · 09/09/2026

...bullshit conservatives only partially succeeded in making a thing starting several years back. It was never about fairness. Regardless, in my opinion, trans athletes in youth sports are so uncommon and variable in skill level that any concerns regarding "fairness" are best handled on a...

100

Bruce Stagg Martin @bstaggmartin.bsky.social · 09/09/2026

Also worth pointing out that, as I understand, many sports regulatory boards already had policies in place (required wait times on HRT, etc.) before the modern "blowup" of these issues. It's a wedge issue trying to ignite a culture war against trans folks, just like the bathroom...

110

Bruce Stagg Martin @bstaggmartin.bsky.social · 09/09/2026

Nice! Regardless of where you stand, I do think we were due for some "rebalancing" of perspectives on this topic in the literature.

010

Reposted by Bruce Stagg Martin

Dan MacGuigan @dmacguig.bsky.social · 08/09/2026

New paper in Systematic Biology! We show how micro-synteny can help resolve a tricky fish #phylogenetic problem where traditional sequence-based analyses fall short. Genomes archive evolutionary history in more than just their basepairs. doi.org/10.1093/sysb... #fish #genomics #evolution

35026

Bruce Stagg Martin @bstaggmartin.bsky.social · 09/09/2026

...provides useful links for further reading. However, I still think solely relying on LLMs tends to encourage superficial understanding with little foundation, and that dedicated study is still ultimately necessary for developing deeper, more accurate intuitions.

101

Bruce Stagg Martin @bstaggmartin.bsky.social · 09/09/2026

More than you'd think with respect to physics/math (though some of that is just for fun rather than work). To be clear, I admit you raise valid points, and I'm just trying to express my perspective. Personally, I've mainly found LLMs useful in the same way Wikipedia is--a good starting point that...

101

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

I think the fact that there's a very real possibility that scientists' use of OpenAI models might have been fed into this proof by accident/negligence really underscores the latter point.

000

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

...to continue scaling up/running them. The ideal situation gets "flipped": with scientists' efforts acting as little more than fuel to continue feeding the AI. I honestly feel this would lead to stagnation with an illusion of productivity, but I again admit I have something of a pessimistic bias.

100

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

...the emphasis placed on "productivity" above all else, I remain unconvinced this is where we're headed. Instead, it seems like we're headed towards a world where LLM outputs are overvalued all while the cutting-edge models lie in the hands of those corporations large enough...

100

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

2) I think you'd be right if we lived an ideal world where scientists always (and will continue to be able to) easily wield these LLMs with their "intellectual sovereignty" intact. However, between the immense costs of these models, excitement sorrounding job automation, and...

100

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

1) I think the "friction" that comes from working through fields with which you are unfamiliar is a crucial source of scientific creativity and insight. AI can help with this for sure, but it can also be used as a crutch that seems to let you skip this process entirely.

200

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

I would contend that the experience of me and most of my colleagues (granted we're biologists/statisticians) is that our most exciting discoveries generally come "by accident" while working through other problems. Two other points:

100

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

It's a good point, and you're right--mathematicians won't be satisfied. I'm not worried about them; I'm worried about funding agencies, etc. We already live in a world where metrics meant to be rough proxies of quality/progress get mistaken for literal quality/progress.

011

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

I also tend to agree with Terence Tao's recent take that there's a subtle danger in this style of mathematics--that it's too focused on end results and circumvents the important knlowedge and insights mathematicians generate on the road to these proofs. Again, might just be my bias though 🤷

2180

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

It will be interesting to see where the dust settles--MAJORLY impressive, but I still have my reservations that these models' strength lies in there ability to combine disparate literature and brute force things rather than make "intuitive leaps". But I admit that could be a bias.

280

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

Navier-Stokes solution is HUGE, no matter how much compute it took. That being said, I do find it hilarious that they completely omitted units from their x-axis here. Makes me wonder how much time/resources are ultimately needed for this level of performance.

290

Bruce Stagg Martin @bstaggmartin.bsky.social · 08/09/2026

There are probably healthy ways to incorporate LLMs into mathematical research, and the current bullshit going on ain't it.

011

Bruce Stagg Martin @bstaggmartin.bsky.social · 07/09/2026

Unfortunately coincident with my annual review deadline--thanks for the heads up! Didn't catch this.

010

Reposted by Bruce Stagg Martin

Sarah McAnulty, Ph.D. @sarahmackattack.bsky.social · 03/09/2026

Attention Philadelphia! We're hosting a rally to demand that Drexel University maintain The Academy of Natural Sciences Museum open to the public. Join us on Tuesday morning at 10am in front of the academy. Please spread the word, we need all hands on deck! RSVP here: luma.com/1qwnrqz2

A dinosaur on a white background with the text SAve the academy of Natural sciences museum rally Tuesday Sept 8 10am in front of the academy 1900 ben Franklin Parkway. "Tell Drexel University "Keep the museum open!" with the logos for Philly Science Action, Skype a Scientist, and Green Philly

171410895

Bruce Stagg Martin @bstaggmartin.bsky.social · 04/09/2026

Ugh, I hear ya--thank god for logs and LogSumExp shenanigans, but then you gotta do numerical linear algebra or Fourier transforms and you're fucked 😭

010

Reposted by Bruce Stagg Martin

John Wares @johnwares.bsky.social · 04/09/2026

I just challenged my local (currently renovating and restructuring) Georgia Museum of Natural History to have an event on September 8 in solidarity. What other NHMs will join? @friel.bsky.social please share in bsky museum world...

2187

Bruce Stagg Martin @bstaggmartin.bsky.social · 04/09/2026

Oh my god yes! I've never seen anyone else talk about the whale skeleton smell--I did some morphometrics work on cetacean skulls in undergrad and had to go to the Smithsonian's off-site marine mammal collection in Suitland, MD. That smell is something else! Such oily bones!

030

Bruce Stagg Martin @bstaggmartin.bsky.social · 28/08/2026

Wow, what a creative and original argument...not like Amazon uses the exact same rhetoric or anything like that. 🙄

000

Bruce Stagg Martin @bstaggmartin.bsky.social · 27/08/2026

Well if your goal is to estimate census population size, I guess effective population size itself could be deemed ineffective.

010

Reposted by Bruce Stagg Martin

Ologies with Alie Ward @ologies.bsky.social · 27/08/2026

Okay, Atlanta: tell me your Ologists I should meet.

9253

Bruce Stagg Martin @bstaggmartin.bsky.social · 21/08/2026

Really fascinating rabbithole! Saw the headline from your 2023 paper but never got a chance to dig into the details before. Any opinions as to what extent the apparent lack of sampled ancestors is an ascertainment and/or phylogenetic inference bias? Seems like David has thoughts. 😬

010

Bruce Stagg Martin @bstaggmartin.bsky.social · 23/07/2026

Again, machine learning is super crucial when a model is otherwise intractable, don't get me wrong! I just find much more joy in explicitly thinking through the model, likelihoods, etc.

020

Bruce Stagg Martin @bstaggmartin.bsky.social · 23/07/2026

THANK YOU! As important as it is, I've always found machine learning somewhat unsatisfying personally...it feels like adding parameters and data until the model does what you want, rather than really carefully thinking through the model's structure and how to efficiently fit it to data.

120

Bruce Stagg Martin @bstaggmartin.bsky.social · 05/07/2026

Sometimes you just get those questions that send you down the most interesting little rabbit holes...

010

Reposted by Bruce Stagg Martin

Eleanore Ritter (she/her) @eleanoreritter.bsky.social · 29/06/2026

Good morning, #SMBE2026! To purge, or not to purge, *that* is the question 🧬 I'll be giving a talk on how mating system transitions shape transposable element dynamics in monkeyflowers this morning in S24 at 11:20 AM - come for the Hamlet pun, stay for the cool comparative genomics!

A dirt road alongside a rocky meadow of yellow flowers and a stand of dead trees

094

Reposted by Bruce Stagg Martin

Tracy Heath @trayc7.bsky.social · 23/06/2026

I’m recruiting a PhD student to join our lab in the fall of 2027. If you’re excited about phylogenetics, comp bio, paleobiology, stats theory, macroevolution, etc. please get in touch! Folks at #Evol2026, you can talk to @brpetrucci.bsky.social & @sigournie.bsky.social to learn more about our lab

08097

Bruce Stagg Martin @bstaggmartin.bsky.social · 22/06/2026

"The AI bubble is about to pop! Be good scientists for the party and find our next big grift...oooh, I bet it will have quantum in the name! It sounds so futurey."

020

Reposted by Bruce Stagg Martin

NPR @npr.org · 21/06/2026

Mona Khalil died Friday after an Israeli airstrike hit her beachside home two weeks ago. She's credited with creating a conservation movement in southern Lebanon to protect sea turtle nesting grounds. n.pr/4oJ4Npy

n.pr

Leading Lebanese conservationist dies after Israeli airstrike on her home

45977406

Bruce Stagg Martin @bstaggmartin.bsky.social · 29/05/2026

You've gotta be kidding me... "You must publish open access" "Great! Here are the additional publication fees I need covered" "No money, only open access" 🙃 I know this isn't the admin that demanded open access...but still, what the hell?

030

Bruce Stagg Martin @bstaggmartin.bsky.social · 20/05/2026

Ooh, I agree! More of this, please. I guess the main issue is resources as always, but I feel like this addresses pressing needs both for academics looking for jobs as well as students who wouldn't mind being taught by profs with more balanced teaching vs research priorities.

010

Reposted by Bruce Stagg Martin

Penn State Department of Biology @psubiodept.bsky.social · 07/05/2026

🌱 We’re hiring in Plant Biology at Penn State! The Department of Biology at Penn State University Park is searching for a tenure-track Assistant or Associate Professor in any area of plant biology.

12746

Reposted by Bruce Stagg Martin

Andrew Hipp @andrewlhipp.bsky.social · 05/05/2026

Join @amsocplanttaxon.bsky.social for Dr. Lauren Eserman-Campbell (Atlanta Botanical Garden)'s webinar, “How Taxonomists Can Shape Conservation Outcomes” on Friday, June 5 at 12pm ET. Register today, and please share! us06web.zoom.us/webinar/regi...

0187