Michael Landis
LightNews LightNews
banner
landismj.bsky.social
Michael Landis
@landismj.bsky.social
2.7K followers 670 following 140 posts
Evolution, phylogenetics, biogeography, epidemiology, computational biology, statistical models, machine learning, scientific software — 🌵🕰️🌎🎲🖥️📈 — https://landislab.org
Posts Replies Media Videos
landismj.bsky.social
It's back! The Society of Systematic Biologists is running Year 3 of the Mentorship Program!

Find community, support your colleagues, and talk trees. Apply by June 10th, 2025.

Program info: www.systbio.org/mentorship-p...

Application form: forms.gle/QtD22C9dtfRF...

@systbiol.bsky.social
Logo for Society of Systematic Biologists Mentorship Program. Two parallel phylogenies, nothing fancy.
May 19, 2025 at 5:21 PM
landismj.bsky.social
Familiar expression from whiteboard, earlier. Seems about right.
February 12, 2025 at 4:06 AM
landismj.bsky.social
Are you referencing this? From doi.org/10.1007/s107...
In the following years a debate flared up as a reaction to a widely circulated, but never published manuscript by Gareth Nelson. In this paper, he stated that cladograms are not phylogenetic trees but tree diagrams representing patterns of unique characters (Eldredge and Cracraft 1980; Wiley 1981).
December 14, 2024 at 9:43 PM
landismj.bsky.social
Here's the second starter pack of Early Career Evolutionary Biologists: go.bsky.app/j5DX3f

Link to first starter pack: go.bsky.app/MzND7nX

Follow these inspirational folks to support the next generation of evolutionary biologists #evobio

Out of time for now! Will add more names this weekend
Screen shot of starter pack image
December 3, 2024 at 2:47 PM
landismj.bsky.social
🤔
Mural from Missouri History Museum Edge of mural from Missouri History Museum Edge of mural from Missouri History Museum w ivory billed woodpecker?
November 29, 2024 at 6:21 PM
landismj.bsky.social
Hennig's answer to Seuss
three fish in a tree
October 24, 2024 at 3:57 PM
landismj.bsky.social
Phylogenetics featured in caller feedback for Savage Love podcast. Is this what it means to be a mature field? #evolution
Transcript says: Then back to Savage love, Junior says in response to last week's intro about the mice who get blasted with both testosterone and cortisone and then stress themselves out to death, Dan wished there was an easy way to know how evolutionarily diverged. He is from that ridiculous australian mouse, which is actually a marsupial. Good news there is. This website offers a research backed evolutionary divergence time for any pair of taxa. Timetree.org dot for example, humans and antichonus last had a common ancestor about 160 million years ago, around when placental and marsupial mammals diverged. We can also check divergence times to other sexy species like sequentially hermaphroditic clownfish Pulmocentridae, 429 million years ago, or pregnant male seahorses, also 429 million years ago. Or praying mantises that practice nuptial cannibalism much more frequently in captivity than in the wild 686 million years ago. Or the fungus with more than 17,000 sexes. Trichaptum 1270 5 million years ago, or poplars, which keep changing their sex chromosomes 1000, 530 million years ago. Ain't science fun?
October 12, 2024 at 4:20 AM
landismj.bsky.social
Excited to share work on a new phylogenetic biome shift model!

Model lets species gain and lose enabled (fundamental) and established (realized) biome affinities through historical events

A joy to work w/ the intrepid Sean McHugh and Michael Donoghue

www.biorxiv.org/content/10.1...

#evolution
Screen capture of ancestral biome affinity reconstruction for subclade in Viburnum. Pie charts colors indicate enabled affinities for biomes, black centers indicate established affinities for biomes. Diagram of biome shift model (RFBS) events. Panels show how species gain and lose enabled and established affinities through anagenetic and cladogenetic events under RFBS.
September 5, 2024 at 2:03 AM
landismj.bsky.social
Seeing that dagger next to an author's name always worries me

Is this some kind of sick joke paleontologists like to play?
August 23, 2024 at 8:51 PM
landismj.bsky.social
The time spent squinting for typos in squirrelly scientific idioms made from everyday words
August 23, 2024 at 8:21 PM
landismj.bsky.social
Does anyone have a reference for dates on transatlantic Africa -> South America dispersal events in mammals? Are there any mammal dispersals younger than bats at ~18 Ma? 🦇
Wikipedia screenshot: "Molecular sequence data indicate T. brasiliensis's closest relatives are the sister species Sauromys petrophilus of South Africa and Tadarida aegyptiaca of Africa and South Asia. These three species form a clade believed to be about 18 million years old."
August 23, 2024 at 8:14 PM
landismj.bsky.social
Trader Joe's public stance on hard polytomies #evolution
August 11, 2024 at 2:09 PM
landismj.bsky.social
Fábio Mendes' new paper on PhyloJunction is now online at Systematic Biology.

Also! He's starting his lab at LSU this Fall and is recruiting PhD students and postdocs to work on phylogenetics, popgen, models, etc. Reach out to him if interested!

Link: academic.oup.com/sysbio/advan...
August 9, 2024 at 2:40 PM
landismj.bsky.social
New preprint on software to fiddle around w/ deep learning + phylogenetic models.

It's meant to be easy to use. You provide a simulation script (any language) and a config file, phyddle does the rest.

In collaboration with the brilliant Ammon Thompson!

phyddle.org
www.biorxiv.org/content/10.1...
figure showing steps for phyddle pipeline Plot showing performance of phyddle (blue) vs. Bayesian (red) estimates for SIR model under a simple (top) and complex (bottom) scenario. Plot comparing model selection results for phyddle (blue) and AIC (red) for various models of continuous trait evolution.
August 9, 2024 at 2:30 PM
landismj.bsky.social
No continental drift necessary to explain biogeographic disjunction
Painting of dispersal-by-ghost https://news.artnet.com/art-world/art-bites-leonora-carrington-surrealist-survival-kit-2456838
June 12, 2024 at 5:20 PM
landismj.bsky.social
Skull envy
Majestic turtle skull
June 11, 2024 at 9:58 PM
landismj.bsky.social
🛷
Sledders on the hill in front of St Louis Art Museum
February 16, 2024 at 11:48 PM
landismj.bsky.social
Join us June 4-6 for a free phylogenetic biogeography workshop using RevBayes at WUSTL

Details: tinyurl.com/phylobiogeow...

Apply by March 6: forms.gle/3ax6q4GzYFMY...

Instructors: Isaac Lichter Marck, Fábio Mendes, Sarah Swiston, Felipe Zapata + me

Funded by award DEB-2040347

#evolsky
Lysimachia venosum (photo by Ken Wood) Graphical representation of GeoSSE model Photo of Washington University in St. Louis Time-calibrated phylogeny
February 15, 2024 at 3:53 PM
landismj.bsky.social
From Nau (2001): "De Finetti used the notation ‘Pr’ to refer interchangeably to Probability, Price, and Prevision (‘foresight’), and he treated them as alternative labels for a single concept."
February 10, 2024 at 1:41 PM
landismj.bsky.social
This approach by Budd & Mann looks nifty. They cross-link rate-heterogeneity for character change and diversification events, using a shared "tempo" parameter. Tempo is driven by OU-ish process, so time runs fast vs. slow w/ autocorrelation at local scales.

www.biorxiv.org/content/10.1...

#evolsky
Figure shows two plots. Plot A shows how evolutionary tempo converges to same stationary distribution regardless of starting tempo. Plot B shows how tempo converges to different stationary distributions depending on diversification rate.
February 6, 2024 at 2:39 PM
landismj.bsky.social
New preprint by Albert Soewongsono (amazing postdoc) on diffusion-based representations of ClaSSE models, with results to map out relationships between SSE rates and stationary state frequencies, and minimum times-to-stationarity #evolsky

arxiv.org/abs/2402.00246
February 3, 2024 at 12:06 AM
landismj.bsky.social
Diagram shows one worm makes two worms, making worms unstoppable
January 27, 2024 at 4:41 AM
landismj.bsky.social
I like the idea of using model misbehavior (e.g. due to bad theory, bad data, or software bugs) to force revisiting of assumptions

Pielou, E. C. (1981). The usefulness of ecological models: a stock-taking. The Quarterly Review of Biology, 56(1), 17-31.

www.journals.uchicago.edu/doi/pdf/10.1...
January 24, 2024 at 7:05 PM
landismj.bsky.social
The point, line, and plane of Euclidean pastametry
January 21, 2024 at 9:37 PM
landismj.bsky.social
Welcome @canales.bsky.social! Also, enjoying Bedeviled thanks to your rec
October 11, 2023 at 2:31 AM