CellFlow was accepted for publication in the journal Nature Methods

Overview Debates Related claims Topics

Nicheformer is a spatial omics foundation model that predicts neighboring cell gene expression states based on an individual cell's transcriptomic profile

The Computational Health Center at Helmholtz Munich includes approximately 45 research group leaders and 500 to 600 research staff members

Engineered organoid models commonly lack vascular perfusion systems and resident immune cell populations

The automated cell type annotation package CellTypist is built upon logistic regression classifiers

Wolfram Schultz recorded dopamine neuron activity in monkeys receiving rewards to examine neural reinforcement mechanisms

Andrew Barto and Richard Sutton shared the Turing Award for their contributions to computer science and reinforcement learning

Stem cells can be differentiated in vitro into three-dimensional multicellular organoids that replicate architectural features of human organs

Cardiac organoids cultured in vitro can form functional cellular networks that exhibit spontaneous rhythmic beating

Wolfram Schultz, Peter Dayan, and P. Read Montague published a paper in Science demonstrating that dopamine firing patterns mirror temporal difference prediction errors

Andrew Barto and Richard Sutton co-authored a foundational textbook on reinforcement learning

The Chan Zuckerberg Biohub announced a $1.8 billion collaboration with the Department of Energy, the National Institutes of Health, Meta, and Google to open-source multiscale biological data for virtual cell modeling

CellFlow applies flow-matching generative modeling to latent representations to predict cell state trajectories resulting from biological perturbations

Fabian Theis's group uploaded a 100-million-parameter single-cell foundation model trained on 350 million cells across 16 species to Hugging Face

Bo Wang co-authored a Nature paper demonstrating that flawed evaluation metrics caused linear models to appear to outperform nonlinear biological models

The Helmholtz Association is Germany's largest research organization by budget allocation, operating research centers across health, information, and physical sciences

The OpenProblems consortium requires benchmark participants to propose and evaluate multiple standardized performance metrics for single-cell algorithms

Spatial omics technologies profile molecular expression while preserving native spatial coordinates and multicellular local niches within intact tissue sections

The ARC Institute hosted the Virtual Cell Challenge to benchmark machine learning methods on single-cell perturbation prediction tasks

Fabian Theis's research group published the single-cell perturbation prediction model scGen in 2019

Perturb-seq workflows can execute genome-wide CRISPR knockouts across up to 100 distinct cell lines