Geert-Jan Huizing

🎉 New preprint! biorxiv.org/content/10.110… STORIES learns a differentiation potential from spatial transcriptomics profiled at several time points using Fused Gromov-Wasserstein, an extension of Optimal Transport. @gabrielpeyre @LauCan88

New preprint! 🚨 Clustering is a powerful tool in genomics & neurosci, but is contingent on the distance /cost chosen between data features. With @samo_hromadka & @myamada0 we used optimal transport on trees 🌳 to learn underlying distances – fast! arxiv.org/abs/2411.07432 1/8

🥳Bravo @SibylleMarcotte, membre de l'équipe Ockham de notre centre, doctorante à @ENS_ULM, lauréate 🏆du Prix Jeunes Talents France 2024 @FondationLOreal-@UNESCO #ForWomenInScience @AcadSciences @4womeninscience Toutes nos félicitations !!! ©Richard Pak et Clémence Losfeld

"Transformers are Universal In-context Learners": in this paper, we show that deep transformers with a fixed embedding dimension are universal approximators for an arbitrarily large number of tokens. arxiv.org/abs/2408.01367

🚨🚨 AI in Bio release 🧬  Very happy to share my work on a Large Cell Model for Gene Network Inference. It is for now just a preprint and more is to come. We are asking the question: “What can 50M cells tell us about gene networks?” ❓Behind it, other questions arose like: “how can we best learn networks for scRNAseq data?”, “How would we assess them?” “What are foundation models in biology actually learning about the cell and its mechanism?” We try to partially answer these questions and present a new model: scPRINT 💫. 🏃 -> it is for now a pre-print and more is to come but here are some of our results: scPRINT is a transformer model trained on 50M cells 🦠 from the cellxgene database, it has novel expression encoding and decoding schemes and new pre-training methodologies 🤖. We propose to use the specificity of scRNAseq data and define a multi task pre-training composed of expression denoising, bottleneck learning and classification. We also propose a new hierarchical classification method to work with the rich hierarchical ontologies used to label cells in cellxgene. (1/2) x.com/razoralign/sta…

Inspiration for the 3D visuals came from this paper x.com/christophertap…

💻Try it out with our easy-to-use Python package! github.com/cantinilab/sto…

🎉 New preprint! biorxiv.org/content/10.110… STORIES learns a differentiation potential from spatial transcriptomics profiled at several time points using Fused Gromov-Wasserstein, an extension of Optimal Transport. @gabrielpeyre @LauCan88

Come check @VCastin_ ´s work on the smoothness of attention 🔥🔥🔥 Happening now, poster 510 !

🥳 I’m very happy to announce our preprint biorxiv.org/content/10.110… ! scConfluence combines uncoupled autoencoders with Inverse Optimal Transport to integrate unpaired multimodal single-cell data in shared low dimensional latent space. @LauCan88 @gabrielpeyre

Happy to be at NeurIPS! I'll present my work on conservation laws for gradient flows (openreview.net/forum?id=kMueE…) tomorrow at 10:15am during the DL theory oral session (Room R06-R09) and during the follow-up poster session (Poster #900). Joint work w. @gabrielpeyre & @RemiGribonval

How does optimal transport allow us to study dynamical systems? How does it connect to control theory, flow matching, & diffusion models? How is it advancing molecular biology research? Find the answers in our tutorial! Recording, slides, & script under bunne.ch/ot_tutorial/.

🔬Pleased to share our latest preprint, "G-PLIP: Knowledge graph neural network for structure-free protein-ligand bioactivity prediction" 📖: doi.org/10.1101/2023.0… 🤝 Collaborated between @Roche and @EPFL #Research #DrugDiscovery #GeometricDeepLearning

Stay tuned for the next live coding session on twitch 👨‍💻 coming soon @ twitch.tv/zaccharier

🥳🥳 New work: arxiv.org/abs/2309.01213 Implicit Regularization of ResNets towards Neural ODEs w. @PierreMari0n, Yu-Han Wu and @gerardbiau We show: ResNet initialized as discretization of a neural ODE -> such a discretization holds throughout training.

Ah, we forgot to tweet this. @gjhuizing from @institutpasteur (@LauCan88 lab) has won the MLCSB best poster award with their work titled "Paired single-cell multi-omics data integration with Mowgli". #ISMBECCB2023

I have written a very short mathematical summary of the main ingredients of denoising diffusion models. mathematical-tours.github.io/book-sources/o… Most of this is directly borrowed from the amazingly clear slides of @ValentinDeBort1 vdeborto.github.io/project/genera…

@gjhuizing from my team @institutpasteur presenting Mowgli @MlcsbC #ISMBECCB2023 If you want to know more about it check the preprint biorxiv.org/content/10.110…

New paper out : a simple way to have optimal transport with sparse displacements🎆 "Monge, Bregman and Occam: Interpretable Optimal Transport in High-Dimensions with Feature-Sparse Maps" arxiv.org/abs/2302.04065 w. @CuturiMarco and Michal Klein A small thread👇

Next Tuesday at 2023-06-27 18:00 CEST will be another community meeting! @gjhuizing will talk about Mowgli, a method for the integration of paired multi-omics data! Looking forward to seeing everyone there!

How cooperation events in different omics can help to infer molecular mechanisms ? Read our new pre-print ! "Molecular mechanisms reconstruction from single-cell multi-omics data with HuMMuS" biorxiv.org/content/10.110… @LauCan88 @ina_deutschmann - explainer thread 1/6