scEpath

scEpath infers cellular trajectories and transition probabilities from single-cell RNA-sequencing (scRNA-seq) data by computing energy landscapes and constructing probabilistic directed graphs to model cell state transitions.

Key Features:

• Energy landscape calculation: Quantifies the energy landscape using 'single-cell energy' and distance-based measures to infer transition probabilities and lineage relationships.
• Probabilistic directed graphs: Constructs probabilistic directed graphs to represent directional transition likelihoods among cell states.
• Pseudotemporal ordering: Generates pseudotemporal orderings to resolve temporal progression and key transition events along trajectories.
• Marker gene identification: Identifies marker genes and gene expression patterns associated with cell state transitions.
• Robustness and flexibility: Demonstrates robustness to variation in input gene set size and operates broadly unsupervised with minimal parameter tuning.

Scientific Applications:

• Developmental trajectories: Reconstruction of developmental trajectories to analyze dynamics of differentiation and cell fate decisions.
• Cell communication networks: Identification of cell–cell communication networks implicated in contexts such as early human embryo development.
• Transcription factor discovery: Discovery of transcription factors involved in processes including myoblast differentiation.
• Lineage dynamics analysis: Analysis of common and branch-specific temporal dynamics and transcriptional programs along branched lineages.

Methodology:

Energy landscape-based inference integrating 'single-cell energy' with distance-based measures to calculate transition probabilities and trajectories, construction of probabilistic directed graphs, generation of pseudotemporal orderings, and identification of marker genes.