Any update on this? Thanks!
09.03.2026 12:29
👍 1
🔁 0
💬 1
📌 0
Any update on this? Thanks!
Delighted to have been a part of this. It's a very exciting area in the field of social evolution and beyond!
Eusociality has independently evolved in multiple arthropod lineages
Eusociality has independently evolved in multiple arthropod lineages
Comparative analysis across 5,678 insect species shows that, when you control for phylogenetic bias, eusociality has not evolved at a faster rate in haplodiploid species. www.pnas.org/doi/10.1073/...
"Our ambitious goals include curing the world of cancer, infectious disease, selfishness, and inclusive fitness theory."
In summary, GLADE provides a flexible, fully phylogenetic way to reconstruct gene content evolution directly from OrthoFinder results.
We hope it will be useful across a wide range of comparative genomics studies!
www.biorxiv.org/content/10.6...
github.com/lauriebelch/...
(10/10)
A phylogeny of 78 mammal species with ant- and termite-eating specialists highlighted, and circles to show where key orthogroups have increased or decreased in size (phylogeny from Upham et al. 2019)
To illustrate what this enables biologically, we applied GLADE across a phylogeny of 78 mammals to investigate convergent genomics changes associated with the evolution of ant- and termite-eating.
(9/10)
Figure showing benchmarking of glade on gene loss events. on two sets of simulated datasets, glade shows high accuracy at identifying loss events, and when it makes errors the errors tend to be small
We benchmarked GLADE on realistic simulated datasets and on real data.
Across datasets, GLADE accurately infers gains, losses, and duplications and reconstructs ancestral gene repertoires with high precision and recall.
(8/10)
GLADE is designed to be simple to run.
It only requires a standard OrthoFinder results directory and outputs branch-specific gains, losses, and duplications, full orthogroup histories, and ancestral genomes at every node.
(7/10)
What makes this especially powerful is what you can do with it.
By placing gene content changes and ancestral genomes on the same phylogeny, GLADE lets you directly connect genomic change with evolutionary patterns across the tree of life.
(6/10)
GLADE reconstructs ancestral gene repertoires alongside the full set of inferred events.
This makes it possible to study gene content evolution in a comparative framework.
(5/10)
Figure showing the workflow for GLADE. GLADE uses an orthofinder results folder, and infers and maps evolutionary events (gains, losses, duplications), as well as reconstructing ancestral gene content
GLADE takes a fully phylogenetic approach.
It uses orthogroups, gene trees, and the species tree to infer gains, losses, and duplications, and to map each event onto the phylogeny.
(4/10)
Inferring when genes were gained, lost, or duplicated is surprisingly challenging.
It requires reconciling gene trees with a species tree, placing events on the correct branches, and remaining robust to tree error and missing data.
(3/10)
Every species’ genome has been shaped by a continual churn of gene gain, loss, and duplication over evolutionary time.
This genomic churn plays a key role in adaptation and diversification across all domains of life.
(2/10)
Calling all OrthoFinder users!
We’ve just released GLADE, a tool to infer gene gains, losses, duplications, and ancestral genomes across a phylogeny.
GLADE runs directly on OrthoFinder results.
www.biorxiv.org/content/10.6...
github.com/lauriebelch/...
(1/10)
FastSpeciesTree: Fast and Scalable Species tree Inference https://www.biorxiv.org/content/10.64898/2026.01.20.700630v1
Very happy to share that I just published a new paper from my thesis! 🎉
We analysed 546 species of ant to understand how extreme specialisation into reproductive and non-reproductive roles evolved.
Key discoveries in thread🧵 👇
Full paper here: academic.oup.com/evolut/advan...
Kissing evolved at least 21 million years ago. www.sciencedirect.com/science/arti... @matildabrindle.bsky.social
The evolutionary and ecological consequences of cooperation
-in American Naturalist by @stuwest.bsky.social, @annadewar.bsky.social, @ryosukeiritani.bsky.social, Laurence Belcher, and @asgriffin.bsky.social
www.journals.uchicago.edu/doi/abs/10.1...
Super interesting session on post-phylogenomics this morning at #eseb2025
I have a poster on the new version of Orthofinder - there will be a suggestion box 🎁
P03.076 on Thursday
Is there a single PDF with a list of talks/times for #ESEB2025 ? I don't get on well with conference apps
Get OrthoFinder here github.com/OrthoFinder/...
Read the preprint here www.biorxiv.org/content/10.1...
Millions of species here we come!
(10/10)
OrthoFinder is not only fast and accurate, it's easy to use
Just provide the complete set of amino acid sequences for your species
If you prefer a specific tree or alignment tool, it's easy to customise
We also provide rich outputs like gene duplications and comparative genomics stats (9/10)
figure showing how orthogroups are now phylogenetically delineated in orthofinder 3
What else is new?
We now use gene tree–species tree reconciliation to refine orthogroups
This catches cases where distinct orthogroups were mistakenly fused (8/10)
figure showing ortholog benchmarking of various orthology inference tools
What about ortholog accuracy?
We tested using the gold standard Quest for Orthologs benchmarking service
OrthoFinder scored highly across the board (7/10)
orthogroup benchmarking figure, using orthobench data
But is it still accurate?
We benchmarked orthogroups using the OrthoBench dataset
OrthoFinder came out on top (6/10)
image showing scalability of orthology inference tools on large datasets
So is it scalable?
We benchmarked OrthoFinder against other widely used orthology tools
OrthoFinder is the only method able to analyse >1000 species within our time cutoff (5/10)
Our trick: run OrthoFinder on a small subset of species first
Next, we sample representative sequences from each orthogroup to build profiles
Genes from new species are then matched to these profiles to assign them to orthogroups
We avoid the costly all-vs-all step that kills scalability (4/10)
Most tools rely on all-versus-all comparisons between species
This becomes painfully slow as datasets grow
We needed a better way (3/10)
Millions of species are being sequenced
That’s a huge opportunity, but also a major challenge
How can we ramp up scalability without compromising accuracy?
That’s exactly what we set out to solve in this update (2/10)
OrthoFinder just dropped a major update
It’s faster, more accurate, and ready for thousands of genomes
Let’s break it down (1/10)
github.com/OrthoFinder/...
www.biorxiv.org/content/10.1...