cellDancer - Estimating Cell-dependent RNA Velocity

cellDancer is a modularized, parallelized, and scalable tool based on a deep learning framework for the RNA velocity analysis of scRNA-seq proposed by Li et al. (Nature Biotechnology, 2023). RNA velocity reflects the dynamic process of cell state transitions based on scRNA-seq experiment data. cellDancer enables the prediction of RNA velocity and the dynamic kinetics of RNA in single-cell resolution.

cellDancer’s key applications

• Enable accurate inference of dynamic cell state transitions in heterogeneous cell populations.

Example: Mouse gastrulation erythroid maturation.

• Estimate cell-specific transcription (α), splicing (β) and degradation (γ) rates for each gene and reveal RNA turnover strategies.

Example: Mouse hippocampus development.

• Improves downstream analysis such as vector field predictions.

Example: Mouse pancreatic endocrinogenesis.

Latest news

• Our work of cellDancer has been published at Nature Biotechnology! (4/3/2023)

• cellDancer has been released to PyPI (3/21/2023).

Support

Welcome bug reports and suggestions to our Github issue page!

Table of contents

cellDancer

• About cellDancer
• API
  • celldancer.adata_to_df_with_embed
  • celldancer.to_dynamo
  • celldancer.export_velocity_to_dynamo
  • celldancer.velocity
  • celldancer.compute_cell_velocity
  • celldancer.pseudo_time
  • celldancer.embedding_kinetic_para
  • celldancer.cdplt.scatter_gene
  • celldancer.cdplt.scatter_cell
  • celldancer.cdplt.plot_kinetic_para
  • celldancer.cdplt.PTO_Graph
  • celldancer.simulation.simulate
• Release Notes
• Frequently asked questions

Tutorials

• Installation
• Data Preparation
  • Preprocessing
  • Transferring adata format to dataframe
• Case study 1: Gastrulation erythroid maturation
  • Import packages
  • Load data
  • Estimate RNA velocity for sample genes
  • Visualize the phase portraits
  • Project the RNA velocity onto the embedding space
  • Estimate pseudotime
  • Display the abundance of spliced RNA along pseudotime
• Case Study 2: Mouse hippocampus development
  • Import packages
  • Load data
  • Estimate RNA velocity for sample genes
  • Visualize the phase portraits
  • Project the RNA velocity onto the embedding space
  • Estimate pseudotime
  • Display the abundance of spliced RNA along pseudotime
  • Visualize the reaction rates, the abundance of unspliced RNA, and the abundance of spliced RNA on embedding space
• Case study 3: Pancreatic endocrinogenesis
  • Import packages
  • Load cellDancer result
  • Project the RNA velocity onto the embedding space
  • Build UMAP based on reaction rates
• Case study 4: Human embryo glutamatergic neurogenesis
  • Import packages
  • Load cellDancer result
  • Visualize the phase portraits
  • Project the RNA velocity onto the embedding space
  • cellDancer is insensitive to the methods of neighbor cells detection
  • Estimate pseudotime

Application of dynamo

• Downstream analysis using Dynamo
  • Import packages
  • Input the prediction of cellDancer to dynamo
  • Project the RNA velocity onto the embedding space
  • Learn the vector fields
  • Jacobian analysis