VeloVAE RNA Velocity Model Performance Comparisons

Encompassing performance comparisons for the VeloVAE model, a Bayesian method for RNA velocity inference from single-cell gene expression data. The model was evaluated on complex biological systems including hematopoiesis, induced pluripotent stem cell reprogramming, the developing mouse brain, and the entire mouse embryo.

Use Cases

• Compare inferred differentiation directions from VeloVAE against previous RNA velocity methods.
• Analyze the accuracy of transcription rate estimation across different biological systems like the developing mouse brain.
• Evaluate the performance of latent time inference against pseudotime methods using manually chosen cell subsets.
• Model gene expression dynamics in complex systems such as the entire mouse embryo using kinetic rate parameters.

Strengths

• Model tested on multiple complex biological systems including hematopoiesis and the entire mouse embryo.
• Performance evaluated on key metrics: data fit, accuracy of inferred differentiation directions, and transcription rate estimation.
• Model incorporates prior distributions on rate parameters and time points and can model lineage bifurcations.

Limitations

• Dataset is very small at 5.5 KB, indicating limited scope, likely containing summary results rather than raw data.
• Specific row count, column structure, and sample data are unavailable, limiting detailed analysis.
• Data format is XLS, which may not be optimal for large-scale computational analysis common in bioinformatics.

Provenance

Source

figshare, author Yichen Gu.

Collection Method

Results from computational modeling and evaluation of the VeloVAE Bayesian inference method.

Freshness

Last updated March 20, 2026.