Velorama

Velorama infers causal gene regulatory networks (GRNs) by integrating single-cell RNA velocity and pseudotime to quantify transcription factor (TF) dynamics during cellular differentiation.

Key Features:

• Directed Acyclic Graph Representation: Constructs a directed acyclic graph (DAG) of single-cell differentiation dynamics using pseudotime or RNA velocity to represent branching trajectories.
• Causal Inference in GRNs: Infers causal relationships between transcription factors (TFs) and target genes by linking dynamic expression changes to regulatory effects.
• Speed of Transcription Factor Interactions: Estimates and quantifies the speed at which TFs influence target genes, relating regulatory function to interaction kinetics.
• Nonlinear Trajectory Accommodation: Accommodates complex, branching differentiation trajectories without requiring cells to be ordered linearly along a pseudotemporal axis.

Scientific Applications:

• Human corticogenesis: Applied to human corticogenesis to analyze TF dynamics during cortical development.
• Disease linkage: Identifies slow TFs associated with gliomas and fast TFs associated with neuropsychiatric diseases.
• Investigating differentiation and disease drivers: Uncovers causal drivers of differentiation and disease by integrating dynamic expression data into GRN analysis.

Methodology:

Constructs a DAG of single-cell differentiation dynamics from pseudotime or RNA velocity and estimates the speed of transcription factor influence on target genes to infer causal TF–target relationships.