Happy to share STORIES out now on Nature Methods
STORIES learns cell fate landscapes from spatial tramscripromics data profiled at several time points, thus allowing prediction of future cell states.
Led by Geert-Jan Huizing and Jules Samaran
www.nature.com/articles/s41...
@pasteur.fr
I'm presenting #CIRCE, a Python package to infer co-accessible DNA region networks 🧬
Based on #Cicero 's algorithm (Pliner et al.), it runs ~150x faster, processing an atlas of 700k cells in less than 40 min! ⛷️
Short paper: doi.org/10.1101/2025...
Code: github.com/cantinilab/CIRCE
1/5 ⬇️
It was received quite enthusiastically here so time to share it again!!!
Our #ICLR2025 blog post on Flow Matching was published yesterday : iclr-blogposts.github.io/2025/blog/co...
My PhD student @annegnx.bsky.social will present it tomorrow in ICLR, 👉poster session 4, 3 pm, #549 in Hall 3/2B 👈
Excited to see scPRINT published in @natcomms.nature.com ! scPRINT is a large cell model for the inference of gene networks pre-trained on 50Mcells with innovative pretraining tasks inspired from biology. Work led by @jkobject.com
I am super happy to annonce that after 9 months under review the first paper of my PhD: scPRINT is finally available on Nature Comms! 🎉🧬
www.nature.com/artic...
AI for life sciences day 2 started!
@m-albert.bsky.social is now teaching AI for image analysis. Felipe Llinares from Bioptimus will then talk about LLMs, with short talks from @jkobject.com from my team and @emordret.bsky.social from @audeber.bsky.social team.
@pasteur.fr @pasteuredu.bsky.social
Very happy to start this new journey!!
I wrote a summary of the main ingredients of the neat proof by Hugo Lavenant that diffusion models do not generally define optimal transport. github.com/mathematical...
Anne Gagneux, Ségolène Martin, @quentinbertrand.bsky.social Remi Emonet and I wrote a tutorial blog post on flow matching: dl.heeere.com/conditional-... with lots of illustrations and intuition!
We got this idea after their cool work on improving Plug and Play with FM: arxiv.org/abs/2410.02423
Johnson-Lindenstrauss lemma in action:
it is possible to embed any cloud of N points from R^d into R^k without distorting their respective distances too much, provided k is not too small (independently of d!)
Better: any random Gaussian embedding works with high proba!
Looking for a postdoc in comp bio, single cell, ml and data science in an amazing city like Paris? 🗼
Come join the Cantinilab!
emploi.cnrs.fr/Offres/CDD/U... (better english version coming up soon)
