Meriem Bensouda, Ph.D. (@bensoudameriem) 's Twitter Profile
Meriem Bensouda, Ph.D.

@bensoudameriem

ML scientist @AltosLabs, PhD @ENS_ULM | AI to find novel molecular mechanisms of cell health | I tweet about AI in biology, target discovery, & immunology

ID: 1062676279491727360

calendar_today14-11-2018 11:59:03

338 Tweet

365 Followers

1,1K Following

Hani Goodarzi (@genophoria) 's Twitter Profile Photo

Congratulations Vevo Therapeutics team! It turns out mapping billions of drug-gene interactions in vivo gives you the right kind of data for training in silico models of humans cancers.

dana_peer (@dana_peer) 's Twitter Profile Photo

This is truly brilliant work. Calling it a foundation model is a stretch (a mistake many people in the field make). But, the only single-cell AI model in its class that I think is worth the GPU energy expended and one that I excited to utilize. My stamp of approval.👍

Greg Brockman (@gdb) 's Twitter Profile Photo

During my sabbatical, one project I really enjoyed was training DNA foundation models with the team at Arc Institute. I've long believed that deep learning should unlock unprecedented improvements in medicine and healthcare — not just for humans but for animals too. I think this

During my sabbatical, one project I really enjoyed was training DNA foundation models with the team at <a href="/arcinstitute/">Arc Institute</a>.

I've long believed that deep learning should unlock unprecedented improvements in medicine and healthcare — not just for humans but for animals too. I think this
Yang Xiao (@yangxiaobme) 's Twitter Profile Photo

Shout out to Yuqi Tan Ph.D. 🎉 The intelligent brain behind spatial proteomics: from a deck of fluorescent images to tabular expression matrix of proteins with spatial coordinates! Check her brilliant work “SPACEc” here: biorxiv.org/content/10.110…

Endpoints News (@endpts) 's Twitter Profile Photo

"Make your algorithms better, your models better. You do have enough data — if you were innovative enough on your algorithm side.” Nobel Prize winner Demis Hassabis where the biotech industry is going with AI, and where it's falling short. endpts.com/isomorphic-lab…

Adam MacLean (@adamlmaclean) 's Twitter Profile Photo

Transition paths across the EMT landscape are dictated by network logic (For explainer 🧵, 👉🟦) biorxiv.org/content/10.110…

Andreas Tiffeau-Mayer (@andimscience) 's Twitter Profile Photo

How to learn generalizable rules across complex sequence-function maps? 💡In our new preprint we propose a framework for learning two-point statistics and use it to discover biophysical rules of TCR specificity that generalize to unseen ligands ! ✨ arxiv.org/abs/2412.13722

How to learn generalizable rules across complex sequence-function maps? 💡In our new preprint we propose a framework for learning two-point statistics and use it to discover biophysical rules of TCR specificity that generalize to unseen ligands ! ✨ arxiv.org/abs/2412.13722
Andreas Tiffeau-Mayer (@andimscience) 's Twitter Profile Photo

Our work on contrastive learning of T cell receptor representations is now out in Cell Systems! Give SCEPTR embeddings a try for your TCR analysis applications: sceptr.readthedocs.io/en/stable/

Valentine Svensson (@vallens) 's Twitter Profile Photo

A pre-trained SCVI model of our Tahoe-100M cell atlas is now available as a HubModel! It can be used for gene expression visualization, differential expression, or cell clustering. Available at Hugging Face with instructions on how to load it (see thread)

A pre-trained SCVI model of our Tahoe-100M cell atlas is now available as a HubModel! It can be used for gene expression visualization, differential expression, or cell clustering. Available at Hugging Face with instructions on how to load it  (see thread)
Brandon White (@bwhite5290) 's Twitter Profile Photo

Announcing Axiom: Eliminating drug toxicity, without using animals! Alex Beatson and I founded Axiom a little over over a year ago with the mission of eliminating drug toxicity by replacing traditional experiments, such as animal testing, with AI models. We are excited to

Sam Rodriques (@sgrodriques) 's Twitter Profile Photo

Extremely excited to announce the recipients of the 2025 FutureHouse Independent Postdoctoral Fellowship! You might already know of them or their work, these are some of the absolute best rising stars in biology research today. Through the fellowship, we will support them in

Extremely excited to announce the recipients of the 2025 FutureHouse Independent Postdoctoral Fellowship! You might already know of them or their work, these are some of the absolute best rising stars in biology research today. Through the fellowship, we will support them in
Anshul Kundaje (anshulkundaje@bluesky) (@anshulkundaje) 's Twitter Profile Photo

I would extend this to applied ML/AI as well. Less focus on bean counting papers & glam status of journals & more focus on making datasets, tools, methods, analyses that have long-lasting real impact on the fields that we work in.

Maria Brbic (@mariabrbic) 's Twitter Profile Photo

Can we build multimodal models by simply aligning pretrained unimodal models with limited paired data? We introduce STRUCTURE 🏗️: a lightweight, plug-and-play regularizer that preserves latent geometry to align frozen unimodal models using <1% of paired data typically used in

Can we build multimodal models by simply aligning pretrained unimodal models with limited paired data? 

We introduce STRUCTURE 🏗️: a lightweight, plug-and-play regularizer that preserves latent geometry to align frozen unimodal models using &lt;1% of paired data typically used in
Patrick Hsu (@pdhsu) 's Twitter Profile Photo

Today Arc Institute releases State, our first perturbation prediction AI model and an important step towards our goal of a virtual cell State is designed to learn how to shift cells between states (e.g. “diseased” to “healthy”) using drugs, cytokines, or genetic perturbations

Today <a href="/arcinstitute/">Arc Institute</a> releases State, our first perturbation prediction AI model and an important step towards our goal of a virtual cell

State is designed to learn how to shift cells between states (e.g. “diseased” to “healthy”) using drugs, cytokines, or genetic perturbations