Matthew Hahn
@3rdreviewer.bsky.social
5.1K followers 870 following 480 posts
Population geneticist
Posts Media Videos Starter Packs
Pinned
3rdreviewer.bsky.social
It's official--we're escaping for a year!

Happy to say that we'll be on leave at the Wissenschaftskolleg in Berlin from August 2025-July 2026.

Looking forward to beer, travel, and scholarly learning ;)
Photo of the Brandenburg Gate
3rdreviewer.bsky.social
Our paper shows that Spinney is much more accurate than using just one tree (usually the species tree), when you want to calculate one rate, test for multiple rates, or even get a more accurate picture of ancestral states.

Try it and tell us what you think!

3/3
3rdreviewer.bsky.social
The software is called Spinney (which is a small group of trees) and it calculates likelihoods by marginalizing across a small group of gene trees!

2/
3rdreviewer.bsky.social
Anyway, no more uncorrected quantitative statements about very large clades with lots of species!

5/5
3rdreviewer.bsky.social
Altschul et al. provide a weigh to take a phylogenetically weighted average, so that (for instance) the fact that there are so many small rodent species doesn't skew the answer.

I have no idea why they were working on this, but presumably it's related to their work on BLAST

3/
3rdreviewer.bsky.social
Have you ever wanted to answer a question like "do mammals weigh more than reptiles?" or "are male insects bigger than female insects?"

If so, you would presumably get a bunch of measurements and take their average. But biological data is correlated due to phylogeny! So it's not so simple

2/
3rdreviewer.bsky.social
The most important paper in evolutionary biology I'd never heard of:

1/

www.sciencedirect.com/science/arti...
3rdreviewer.bsky.social
Biergarten nummer sieben

(It’s getting cold out, but I’m just going to keep saying “Biergarten”)
3rdreviewer.bsky.social
This is a great opportunity. Apply!
wiko-berlin.bsky.social
Apply online by November 1st! Wiko Early Career Call for fellowships 2026/27 in the #NaturalSciences and #LifeSciences. Gain time to think during a 3–5 month #fellowship at the Wissenschaftskolleg zu Berlin – Institute for Advanced Study: wiko-berlin.de/fellow-werde... #cfa
3rdreviewer.bsky.social
This paper started because @fabiology.bsky.social and I had found that parsimony *was* consistent under the MSC (in 2018), at least for rooted 4-taxon trees. I thought maybe this would always be the true.

But no--Daniel showed definitively this is not the case. Glad that was cleared up!

2/2
3rdreviewer.bsky.social
Happy that our new paper is out, led by super-postbacc Daniel Rickert (now a PhD student at UNC). Also kind of excited to have my first TPB paper...

1/2

"Inconsistency of parsimony under the multispecies coalescent"
www.sciencedirect.com/science/arti...
Inconsistency of parsimony under the multispecies coalescent
While it is known that parsimony can be statistically inconsistent under certain models of evolution due to high levels of homoplasy, the consistency …
www.sciencedirect.com
Reposted by Matthew Hahn
speciationlab.bsky.social
kites over Berlin,
tempelhofer feld
dozens of kites soar in a blue sky, over thousands of people
3rdreviewer.bsky.social
I actually have no idea what study or finding you were even commenting on. Is there a link?
3rdreviewer.bsky.social
It is useful! And so much better with Overleaf than it used to be. Whatever works best for you.
3rdreviewer.bsky.social
Scientists, DON'T USE LaTeX UNLESS ABSOLUTELY NECESSARY, DON'T GIVE LaTeX TO YOUR YOUNG CHILD FOR VIRTUALLY ANY REASON, BREAK UP THE .tex DOC INTO THREE TOTALLY SEPARATE DOCS (NOT MIXED!), TAKE [Underfull \vbox (badness 10000) has occurred]