PulseAugur
EN
LIVE 16:02:02

AI models advance cell dynamics prediction with novel architectures

By PulseAugur Editorial · [4 sources] ·

Researchers are developing new AI models to predict cellular dynamics and responses to perturbations. One approach, Chreode, uses a cell world model for one-step temporal predictions and has shown improvements in gene-state embedding for perturbation prediction. Another line of research explores temporal graph learning, where cellular states are modeled as evolving graph structures, outperforming existing foundation models in forecasting biological states. Additionally, a latent dynamical causal generative model, CITE-VAE, aims to capture latent cellular programs and their perturbation-driven dynamics, demonstrating improved generalization to unseen perturbations. AI

IMPACT These advancements in AI models for biological systems could accelerate drug discovery and personalized medicine by enabling more accurate in-silico predictions of cellular responses.

RANK_REASON Multiple research papers published on arXiv detailing novel AI models for biological system prediction.

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 4 sources. How we write summaries →

AI models advance cell dynamics prediction with novel architectures

COVERAGE [4]

  1. arXiv cs.LG TIER_1 English(EN) · Mufan Qiu, Genhui Zheng, Yinuo Xu, Ruichen Zhang, Ying Ding, Qi Long, Tianlong Chen ·

    Chreode: A Cell World Model for One-Step Temporal Dynamics and Perturbation Prediction

    arXiv:2605.28111v1 Announce Type: new Abstract: Predicting how a cell will change its transcriptional state under a developmental signal or a genetic perturbation is the computational core of in-silico biology and the AI Virtual Cell program. Existing approaches either fit static…

  2. arXiv cs.LG TIER_1 English(EN) · Manuel Dileo, Andrea Sottoriva ·

    Applications of temporal graph learning for predicting the dynamics of biological systems

    arXiv:2605.28659v1 Announce Type: new Abstract: Biological foundation models have shown strong performance in single-cell representation learning by applying transformer architectures directly to gene-expression matrices. However, these approaches predominantly operate in static …

  3. arXiv cs.LG TIER_1 English(EN) · Andrea Sottoriva ·

    Applications of temporal graph learning for predicting the dynamics of biological systems

    Biological foundation models have shown strong performance in single-cell representation learning by applying transformer architectures directly to gene-expression matrices. However, these approaches predominantly operate in static settings and do not explicitly model the tempora…

  4. arXiv cs.LG TIER_1 English(EN) · Wenkang Jiang, Yuhang Liu, Erdun Gao, Ehsan Abbasnejad, Lina Yao, Javen Qinfeng Shi ·

    Learning Latent Dynamical Causal Processes for Single-Cell Perturbation Prediction

    arXiv:2605.25581v1 Announce Type: new Abstract: Single-cell perturbation prediction aims to infer how cells respond to unseen interventions and to achieve out-of-distribution (OOD) generalization, providing a computational route to understanding how perturbations reshape cellular…

RELATED ENTITIES

RELATED TOPICS