VeloViz

VeloViz constructs RNA-velocity-informed 2D and 3D embeddings of single-cell transcriptomic data to visualize cellular trajectories and predicted future transcriptional states.

Key Features:

• RNA-Velocity Integration: Leverages RNA velocity analysis to predict future transcriptional states and incorporate those predictions into embeddings.
• Embedding Consistency across Topologies: Generates 2D and 3D embeddings that capture cellular trajectories and preserve dynamic relationships even when intermediate cell states are absent.
• Validation on Real and Simulated Data: Demonstrated robustness on both real and simulated single-cell transcriptomic datasets for representing complex biological processes.
• Comparison to PCA and t-SNE: Addresses limitations of PCA and t-distributed stochastic neighbor embedding (t-SNE) by integrating predicted transcriptional dynamics rather than relying solely on observed transcriptional states.

Scientific Applications:

• Single-cell trajectory analysis: Visualizes dynamic gene expression changes to support analysis of differentiation pathways in single-cell RNA-seq data.
• Lineage tracing and cell state transitions: Aids interpretation of lineage relationships and transitions between cell states using RNA-velocity-informed embeddings.
• Developmental biology and cancer progression: Facilitates study of developmental processes and tumor evolution by representing temporal and directional aspects of gene expression.

Methodology:

Integrates RNA velocity predictions with traditional single-cell transcriptomic data to inform construction of 2D and 3D embeddings that reflect current and predicted future transcriptional states.