ssPATHS

ssPATHS generates pathway scores from expression data for single samples after training on a reference cohort.

Key Features:

• Single-sample pathway scoring: Produces pathway scores for individual samples rather than requiring cohort-level analysis.
• Reference cohort training: Trains on a reference cohort to establish pathway activity baselines.
• Linear discriminant analysis: Performs linear discriminant analysis on gene set expression to distinguish samples with augmented versus non-augmented pathway activity.
• Hyperplane-based scoring: Uses the separating hyperplane from the discriminant analysis to score new samples.

Scientific Applications:

• Pathway activity inference: Quantifies the activity level of biological pathways in individual samples based on gene expression data.
• Cross-cohort analysis: Enables comparison of pathway scores between new samples and a reference cohort.

Methodology:

The method takes the expression of a gene set (pathway) from a reference cohort and performs linear discriminant analysis to distinguish samples that have the pathway augmented from those that do not. The separating hyperplane is then used to score new samples.