New OpenFold3 preview out! (OF3p2)
It closes the gap to AlphaFold3 for most modalities.
Most critically, we're releasing everything, including training sets & configs, making OF3p2 the only current AF3-based model that is functionally trainable & reproducible from scratch🧵1/9
Posts by Guillaume Ramstein
(1/2) Interested in plant evolution? We are opening two PhD positions in my lab at QGG - Aarhus University, where you will combine comparative genomics and machine learning to better understand and improve crop traits.
#PlantSciJobs #evolution #genomics
Genetic background shapes AI-predicted variant effects www.biorxiv.org/content/10.64898/2026.04...
GxE interactions were documented in 264 ryegrass accessions at Nordic/Baltic distribution limits. Accounting for GxE improved prediction accuracy up to 22% for spring cover. GWAS identified large-effect variants with environment-dependent adaptive effects for persistence traits.
🎓 Applications open for Master 2 EvoGEM which provides theoretical and methodological training to interpret evolutionary patterns from genomic data.
✔️ Evolution, Maths, Population and quantitative genetics, Phylogenetics
✔️ Tutored projects
✔️Fully taught in English
📍Paris
evogem.fr
📢 Please share!
Have we been overstating the role of transposable elements in adaptation to local or rapidly changing environmental conditions? Happy to share my (somewhat unpopular?) opinion paper on this matter: www.sciencedirect.com/science/arti...
Thanks to those who already shared the link!
Why do complex traits differ in their genetic architecture?
In our new PLOS Biology paper, we will try to convince you that two simple scaling laws drive differences in the number, effect sizes and frequencies of causal variants affecting complex traits.
Thread:
journals.plos.org/plosbiology/...
We are excited to share GPN-Star, a cost-effective, biologically grounded genomic language modeling framework that achieves state-of-the-art performance across a wide range of variant effect prediction tasks relevant to human genetics.
www.biorxiv.org/content/10.1...
(1/n)
Out after peer review, collaborative study from Nordborg & Weigel labs with help from many others. Not the largest collection of new Arabidopsis thaliana genomes, but we hopefully put forward some good ideas for how to think about pangenomes and their analysis!
www.nature.com/articles/s41...
Happy to share this review led by Janek Sendrowski, and written with Thomas Bataillon. We provided an overview of sequence-based AI models: what they can (and cannot) do, and what differentiates them from traditional techniques based on statistical associations link.springer.com/article/10.1...
Interested in using functional data to understand clinical variants? Been hunting for a good review of the topic? We just wrote one! rdcu.be/exaEU
Our work on "Evaluating the representational power of pre-trained DNA language models for regulatory genomics" led by @AmberZqt with help from @NiraliSomia & @stevenyuyy is finally published in Genome Biology! Check it out!
genomebiology.biomedcentral.com/articles/10....
This work started nearly 10 years ago and was once my main postdoctoral project at @plantevolution.bsky.social before I slowed work on it to a trickle because it became confusing. But it always remained extremely interesting.
journals.plos.org/plosbiology/...
What a community, what a welcoming and inclusive atmosphere, what a location! Thank you @snsf.ch, CSF @ethz.ch, @newphyt.bsky.social, @biologists.bsky.social and @cerealcell.bsky.social for supporting us.
🌾Brachypodium 2025 was a blast🌾
Grazie mille - Thank you so much
Forza Brachy - Go Brachy
Nature research paper: A haplotype-resolved pangenome of the barley wild relative Hordeum bulbosum
go.nature.com/45V5IvM
Aarhus University seeks a tenure-track assistant professor in population genetics, comparative genomics, or bioinformatics for biodiversity research. More info: www.au.dk/om/stillinger/job/tenure... #job
Excited to launch our AlphaGenome API goo.gle/3ZPUeFX along with the preprint goo.gle/45AkUyc describing and evaluating our latest DNA sequence model powering the API. Looking forward to seeing how scientists use it! @googledeepmind
Nearing 400 members on our Zeavolution slack! Want to interact with other members of the maize genetics community? We have a weekly webinar on maize diversity/evolution as well. DM me and I'll share a link to join the slack.
Our summer course on #genomicPrediction with me and @guillramstein.bsky.social is running this year 4-8 August, application deadline 18 March. Info and registration see: international.au.dk/education/ad...
First slide of the maize meeting.
Travel permits of all US government employees (USDA, NSF) have been revoked shortly before the the maize meeting affecting:
- The chair of the meeting
- The winner of the most prestigious McClintock award
- A plenary speaker
- Two short talk presenters
- 22 scientists presenting posters.
#MGM2025
🚨New paper published in @nature.com! Using pan-genetics across the Solanum genus🍅🥔🍆we reveal why gene duplications🧬are major contingencies in crop engineering. My postdoc work in the Lippman lab @CSHL, collab. with @katiejenike.bsky.social @mikeschatz.bsky.social chatz.bsky.social and many others!
I am happy to announce our new paper "Univariate-guided sparse regression". It's a new lasso that leverages the signs and magnitude of univariate coefficients .
Sparser and more interpretable than the lasso. We're excited! arxiv.org/abs/2501.18360
R: github.com/trevorhastie...
Please feel free to reach out to Torben Asp or DM me if you have any questions regarding the offer.
The Center for Quantitative Genetics and Genomics (QGG) in Denmark is looking for talented candidates who have interests in quantitative genetics or related fields. The working environment at QGG is great, and Denmark has very good opportunities for research funding.
tech.au.dk/en/about-the...
PhD position on the domestication of common beans and its coevolution with rhizobia through studying ancient and modern genomes. At U. Copenhagen with
Jazmín Ramos Madrigal.
candidate.hr-manager.net/ApplicationI...
On top, five Andropogoneae grass species, from left to right Zea mays subsp. mays, Hemarthria uncinata, Miscanthus sinensis, Sorghum bicolor, Andropogon gerardi. On bottom, a map of the world, with collection sites of Andropogoneae grasses in points. A green background shows the distribution of Andropogoneae grasses throughout tropical and temperate latitudes
On left, a phylogeny of Andropogoneae species, with 14 independent polyploidy origins noted. On right, genome size and repeat content of each assembly. Across all assemblies, average assembly size is 1.9 Gb, and average repeat size is 1.5 Gb.
New preprint featuring my favorite maize-relatives, the Andropogoneae! We sequenced the genomes of 27 species, including lemongrass, kangaroo grass, little and big bluestem, and more -- key species that shape grasslands and prairies worldwide. www.biorxiv.org/content/10.1...
I totally forgot that in addition to the radical genomics, Michelle put together a website where you can learn about the biology of all the species we looked worked with: mcstitzer.github.io/panand_assem...
The consequences of polyploidy are... well, complicated. A deep dive into polyploidy in a massively successful tribe of grasses. A masterpiece led by Michelle Stitzer, representing the work of a lot of folks on the PanAnd project over last 7 years. www.biorxiv.org/content/10.1...
with a proper link to the paper...
nph.onlinelibrary.wiley.com/doi/full/10....
New podcast!
Meet Heredity editor Bastiaan Star (University of Oslo), we talk about his work with ancient fish DNA and why he enjoys working with the journal
@gensocuk.bsky.social @biovitenskap.bsky.social