Comparison between a conventional tool and deep learning models for RNA velocity analysis of scRNA-Seq data

Clinical Snapshot

Key Findings

• P Value: Not specified

• Effect Size: Significant improvement in directional concordance and trajectory continuity

• Primary Outcome: VAE-based methods produced more directionally coherent and consistent velocity fields than classical scVelo

• Nnt Or Sensitivity: Cosine similarity and mean squared error metrics used for performance assessment

• Confidence Interval: Not reported

Clinical Application

Requires computational expertise and resources for deep learning implementation Relevant to Australian research institutions and bioinformatics facilities; no specific regulatory considerations as this is a computational methodology study Researchers conducting single-cell RNA sequencing analysis in any biological context

Abstract

Single-cell RNA sequencing (scRNA-Seq) enables analysis of gene expression at single-cell resolution. RNA velocity analysis infers the temporal dynamics of transcriptional states from the relative abundances of spliced/unspliced mRNA quantified via scRNA-Seq. Classical RNA velocity approaches, such as scVelo, implement gene-specific kinetic modeling. Deep learning methods including DeepVelo, VeloVI, LatentVelo, SymVelo, and scTour are based on variational autoencoders (VAEs), which allow to enhance the robustness and accuracy by leveraging nonlinear latent representations. Here, we systematically evaluated the performance of deep learning RNA velocity tools by comparing with the scVelo dynamical model to access the possible advantages of VAE-base methods. For this purpose, public datasets (GSE149689 and GSE203233) were initially processed using a standard scRNA-Seq pipeline. Comparisons among results of selected velocity tools were conducted using cosine similarity of velocity vectors to assess directional concordance, and by mean squared error analysis of trajectory continuity for the deep learning models. Overall, VAE methods produced significant, richer, and more directionally coherent and consistent velocity fields than the classical model. Our findings indicate that deep learning models provide more consistent and biologically plausible cell-state trajectories, although at the expense of higher computational demands and reliance on accurate splicing quantification. Altogether, our results underscore the relevance of VAE-based frameworks to advance RNA velocity analysis while highlighting the need for careful preprocessing.

References

1. 1.Sauda, M. R., Rodrigues, A. B., de Oliveira Lyra, M. L., Grotto, R. M. T., Gu, L., de Campos Melo, T., & Valente, G. T. (2026). Comparison between a conventional tool and deep learning models for RNA velocity analysis of scRNA-Seq data. Molecular Genetics and Genomics. https://doi.org/10.1186/s13059-023-03065-x