Fast Algorithm for the Log P-value of Fisher s Exact Test or Hypergeometric Distribution

Fast Algorithm for the Log P-value of Fisher s Exact Test or Hypergeometric Distribution computes log p-values for Fisher's exact test or the hypergeometric distribution to support rapid enrichment assessment for enrichment-constrained, time-dependent clustering of microarray time-series data exemplified by the Enrichment Constrained Time-Dependent Iterative Signature Algorithm (ECTDISA).

Key Features:

• Time-Dependency Incorporation: Utilizes a sliding time window to account for temporal dependencies among samples in time-series gene expression data.
• Enrichment Constraint Framework (ECF): Imposes biological enrichment constraints on clustering parameters to ensure identified clusters are both statistically significant and biologically meaningful.
• Supervised Identification of Temporal Transcription Modules (TTMs): Facilitates discovery of TTMs, groups of genes with coordinated expression patterns over time.
• Mathematical Rigor: Employs rigorous mathematical definitions as objective functions for retrieving biologically significant modules.
• Iterative Cluster Refinement: Refines clusters iteratively using combined temporal information and biological relevance criteria.

Scientific Applications:

• Gene Expression Analysis: Applied to human time-series microarray data for Kaposi's sarcoma-associated herpesvirus (KSHV) infection in endothelial cells to confirm known biology and uncover new molecular insights.
• Biological Insight Generation: Integrates clustering with enrichment analysis to guide identification of biologically significant clusters and regulatory mechanisms.

Methodology:

Computational steps explicitly include application of a sliding time window to model temporal dependencies, imposition of enrichment constraints on clustering parameters, and iterative refinement of clusters using temporal information together with biological relevance criteria.