Lightnews — Scholar-powered news

Sean A. S. Anderson 🍉

@seananderson.bsky.social

510 followers 900 following 130 posts

Assistant Prof at Georgia Tech. Computational biology 🤝 field biology. Evolutionary ecology 🤝 evolutionary genetics. Thinking about how one species splits into two. https://seanasanderson.github.io/

seanasanderson.github.io

Posts Media Videos Starter Packs

Pinned

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 22d

Friends: I'm recruiting at least one PhD student to join our lab in Atlanta in the Fall of 2026. Information about our research can be found at the link below. Please pass along to any students interested in the quantitative study of speciation, secondary contact, and the build-up of biodiversity!

The Anderson Lab at Georgia Tech

my academic website

seanasanderson.github.io

20 12

Reposted by Sean A. S. Anderson 🍉

Julia Kreiner @jmkreiner.bsky.social · 8d

Human-mediated land-use and climate change occur simultaneously, but how do they interact to shape adaptive dynamics? Super excited to share the first paper from the Kreiner lab, led by postdoc extraordinaire @rpineau.bsky.social

3 17 55

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

Thanks Jesper!

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

We hope you find this paper and the phylopairs package useful. Please have a read and get in touch if questions arise.

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

Finally, a major impediment to the comp analysis of LP traits has been a lack of ready-made tools for empiricists to deploy. To help on this front, I created the R package 'phylopairs', a tool entirely devoted to working with LP traits. Phylopairs encodes our method plus previous approaches

phylopairs: Comparative Analyses of Lineage-Pair Traits

Facilitates the testing of causal relationships among lineage-pair traits in a phylogenetically informed context. Lineage-pair traits are characters that are defined for pairs of lineages instead of i...

cran.r-project.org

1 1 3

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

The paper goes into much more detail on our approach and a few other methods that have been used in the past. We also provide the first test of the performance of any of the methods that have been used in comparative analyses of LP traits, some of which turn out to be potentially problematic

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

A major breakthrough for us was stumbling across Isserlis' theorem, which allows us to write the expected covariance among lineage-pairs in terms of the expected covariance among the component lineages (which we know if we have a tree).

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

From this basic assumption, we can derive metrics for the expected covariance among PAIRS of lineages. In effect, we convert a standard phylogenetic covariance matrix into a lineage-pair covariance matrix -- and we can use this to account for non-independence in comp. studies of lineage-pair traits

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

Our basis for these expectations is the idea that C and D are similar in some trait(s) due to shared evolutionary history, and this affects the value of the LP traits measured for AC and AD. That is, we assume phylogenetic signal in some underlying character that affects the LP trait we care about

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

Our answer is that, yes, as a NULL expect'n, AC and AD should be more similar to each other in an LP trait than either is to AB. E.g., if A has strong postzygotic RI w/ species C, then it prob. does so w/ D also. If A competes strongly for food w/ C, it prob does so w/ D as well (as a null expect'n)

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

We start by considering the following tree. Six lineage-pairs can be derived from this tree, three of which are AB, AC, and AD. We ask: of those three pairs, would we expect any two to be more similar to each other than either is to the third in some continuous lineage-pair trait?

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

In this paper, we work to develop such models and show that they produce a ‘lineage-pair covariance matrix’, an object that can be used to account for non-independence in downstream analyses in the same that way a ‘phylogenetic covariance matrix’ is used in comparative studies of species’ traits

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

Like the traits of related species, the traits of related lineage pairs are non-independent. Unlike for the traits of species, we currently lack models for the structure of non-independence of lineage-pair traits, which makes it unclear how to account for their underlying covariance.

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

But despite their outsized impact, esp. in the study of speciation, comparative analyses of LP traits have posed statistical challenges that have yet to be fleshed out. Chief among these is non-indendence.

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

This approach has had major impacts since at least Coyne and Orr’s pioneering 1989 comparative study of RI in Drosophila (and that paper's many descendants). Some of the very few ‘general rules’ in evolutionary biology have emerged from these types of analyses.

PATTERNS OF SPECIATION IN DROSOPHILA

To investigate the time course of speciation, we gathered literature data on 119 pairs of closely related Drosophila species with known genetic distances, mating discrimination, strength of hybrid st...

onlinelibrary.wiley.com

1 2

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

A comparative analysis of lineage-pair (LP) traits involves calculating such variables among pairwise combinations of related taxa (not just sisters) and testing for relationships among those traits.

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

First, a ‘lineage-pair trait’ is an attribute that can only be defined for a pair of taxa, NOT a single species or pop'n. So where a species’ trait might be something like ‘mean body mass’, a lineage-pair trait might be something like ‘strength of reproductive isolation (RI)’ or ‘diet niche overlap’

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 1d

Now available in Systematic Biology, a new paper (and R package) in which we outline an approach to account for non-independence in comparative analyses of lineage-pair traits academic.oup.com/sysbio/advan...

The Comparative Analysis of Lineage-Pair Traits

Abstract. For many questions in ecology and evolution, the most relevant data to consider are attributes of lineage pairs. Comparative tests for causal rel

academic.oup.com

2 22 40

Reposted by Sean A. S. Anderson 🍉

Brian O'Meara @omearabrian.bsky.social · 1d

Finally posted blog posts on our latest phylogenetics readings in our modern phylo class:

Phylogeography

Simulation

Behavior & Genomics

Reticulate evolution

All in series: brianomeara.info/blog.html#ca...

#Evolution 🧪

2 14 33

Reposted by Sean A. S. Anderson 🍉

James T. Stroud @jameststroud.bsky.social · 1d

@jblosos.bsky.social and I (and, it turns out, many other lizard biologists!) kept seeing wild lizards with missing legs/hands/feet.

How do they survive? Surely natural selection would weed out these unfortunate individuals?

Perhaps not...🦎🦵

theconversation.com/3-legged-liz...

3-legged lizards can thrive against all odds, challenging assumptions about how evolution works in the wild

Most lizards probably don’t survive devastating injuries. But a new study documents 122 cases of limb loss across 58 species – these exceptions shine a new light on natural selection.

theconversation.com

8 24

Reposted by Sean A. S. Anderson 🍉

Ada Palmer @adapalmer.bsky.social · 7d

I see pieces like this a lot, often w/ a spin of lamenting cultural degeneration, but reading is a LABOR issue, it’s declined because so many people are working overtime or two jobs & employers expect after hours work. France has Earth’s highest reading rate b/c long lunch breaks & labor protections

After Dinner Conversation® @afterdinnerconversation.com · 8d

150 2.9K 8.8K

Reposted by Sean A. S. Anderson 🍉

Brian O'Meara @omearabrian.bsky.social · 12d

Webpage on the hidden curriculum of applying to ecology and evolution grad programs in the US (a lot of the info applies to other fields too, but perhaps less well): applyingtoeeb.info

#AcademicSky 🧪

Applying to US Graduate School in Ecology, Evolution, and Related Fields

applyingtoeeb.info

3 36 51

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 13d

RIP Jane Goodall. There's of course no better way to understand Goodall than in her own words, of which there are volumes to explore. But I also love Stephen Jay Gould's introduction to her "In the Shadow of Man": archive.org/details/insh.... As Gould notes, and Goodall embodied, "Nature IS context"

2 16 37

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 14d

Some great geckos you got there

1 1

Sean A. S. Anderson 🍉 @seananderson.bsky.social · 15d

"Four years on, I see it clearly: the scientific system was never designed for equity. For people whose privilege stems from geography, institutional reputation and inherited networks, the system works perfectly the way it is. That’s why institutions focus only on improved access and optics."

Prof. Dr. Dolors Armenteras @darmenteras.bsky.social · 15d

I can’t promise this will be the last time I talk about inequity — or equity washing — in science. But after the multitude of responses to my Nature piece, I know something has shifted. Many came from the Global North. That alone feels like a small win. 👉 rdcu.be/eGEun

Equity in science is a beautiful lie — and I’m done pretending | Nature

Science isn’t really moving towards equity; institutions are just perfecting the appearance of equity. We need to build an alternative system. Science isn’t really moving towards equity; institutions are just perfecting the appearance of equity. We need to build an alternative system.

rdcu.be

1 3

Reposted by Sean A. S. Anderson 🍉

Justin @ediparolev.bsky.social · 15d

Late Afternoon Light ~ Cyril Cox, 1958

110 1.6K 11K