A latent space thermodynamic model of cell differentiation | Ali Poursina
Join the reading group: https://multiomics-reading-group.gith... Paper: A Latent Space Thermodynamic Model of Cell Differentiation https://www.biorxiv.org/content/10.64... Abstract: Inferring the governing dynamics of differentiation that capture cell state evolution remains a central challenge in single-cell biology. We present Latent Space Dynamics (LSD), a thermodynamics-inspired framework that models cell differentiation as evolution on a learned Waddington landscape in latent space. LSD jointly infers a low-dimensional cell state, a differentiable potential function governing developmental flow, and a local entropy term that quantifies cellular plasticity. Using a neural ordinary differential equation, LSD reconstructs continuous differentiation trajectories from time-ordered single-cell data. Across diverse developmental systems, LSD accurately recovers lineage hierarchies, predicts fate commitment for unseen cell types, and outperforms existing trajectory inference approaches in directional accuracy. Moreover, in silico gene perturbations reveal how individual regulators reshape the landscape, and entropy provides a quantitative measure of plasticity in development and cancer.
Tripso: Self-supervised learning for a gene program-centric view of cell states | Marie Moullet
How AI Cracked the Protein Folding Code and Won a Nobel Prize
Neil Turok’s stunningly simple, testable new theory of the universe
PertEval-scFM: Benchmarking Single-Cell Foundation Models for Perturbation Effect Prediction
LIVE: Conan O’Brien speaks at Harvard graduation ceremony (full)
Generative Modeling via Drifting | MingYang Deng
Prof. Joseph Thywissen - Saturated Collisional Resistivity in Ultracold Hubbard Metals
Learning Adaptive Perturbation-Conditioned Contexts for Robust Transcriptional Response Prediction
"A.I. and Our Economic Future," Professor Chad Jones
High-throughput force spectroscopy of DNA sequence variant libraries | Matt DeJong
Trump sta per invadere l’Iran? La diretta con Alessandro Orsini
The French Do Not Care About Work
AlphaFold - The Most Useful Thing AI Has Ever Done
General Multimodal Protein Design Enables DNA-Encoding of Chemistry
Exposing the LIES of the 20th Century | Aaron Bastani Meets Tariq Ali
Terence Tao: Nobody Understands Why AI Actually Works
scTimeBench: A streamlined benchmarking platform for single-cell time-series analysis
Harnessing AI to Build Virtual Cells | Le Song and Pan Li
The Nature of Reality: A Dialogue Between a Buddhist Scholar and a Theoretical Physicist
