phenopath

phenopath infers pseudotime trajectories from cross-sectional datasets and single-cell gene expression while accounting for heterogeneous genetic, phenotypic, and environmental backgrounds.

Key Features:

• Pseudotime Inference: Extracts latent temporal information (pseudotime) from cross-sectional genomic and gene expression datasets.
• Handling Heterogeneity: Models non-homogeneous genetic, phenotypic, and environmental backgrounds to enable trajectory inference in heterogeneous cohorts.
• Interaction Detection: Identifies interactions between heterogeneous factors and inferred genomic trajectories to reveal effects on gene expression pathways.
• Application to Diverse Data Types: Applicable to single-cell gene expression data and population-level cancer studies.
• R Implementation: Implemented in R.

Scientific Applications:

• Single-cell 'Omics: Analyze cellular differentiation and temporal progression of cell states from single-cell gene expression data.
• Cancer Modelling: Investigate tumour progression and how genetic and environmental factors interact with genomic trajectories in population-level cancer studies.

Methodology:

Employs a statistical framework that learns pseudotime trajectories while accounting for non-homogeneous backgrounds and identifies interaction effects between genetic and environmental factors within gene expression pathways.