AEenrich

AEenrich performs adverse event enrichment analysis to detect biologically meaningful groups of adverse events in vaccine safety studies using AE ontology and enrichment tests.

Key Features:

• AE ontology grouping: Organizes related adverse events into biologically meaningful groups using the AE ontology to improve interpretability of AE signals.
• AEFisher test: Implements an adaptation of Fisher's exact test tailored to pre-selected significant adverse events and AE count data with many zeros and ties.
• AEKS test: Implements a test based on a modified Kolmogorov–Smirnov statistic to accommodate excessive zeros and tied values in AE counts.
• Extension of gene enrichment methods: Extends gene enrichment test concepts to adverse event (AE) count data to enable enrichment-style inference for AEs.
• Simulation validation: Validates the performance of the enrichment tests through extensive simulation studies.
• Application to VAERS: Applies the methods to real Vaccine Adverse Event Reporting System (VAERS) data for practical safety assessments.
• R implementation: Provides the methodology as an R package implementation for computational analyses.

Scientific Applications:

• Vaccine safety signal detection: Identifies enriched groups of adverse events in VAERS to support vaccine safety investigations.
• Biological interpretation of AE signals: Facilitates interpretation of AE signals by grouping events with shared biological meaning via the AE ontology.
• Methodological research on AE data: Enables evaluation and comparison of enrichment methods for AE count data characterized by zeros and ties.

Methodology:

Uses the AE ontology for grouping; extends gene enrichment tests to AE count data; implements AEFisher (adapted Fisher's exact test) and AEKS (modified Kolmogorov–Smirnov statistic); addresses excessive zeros and ties; validated by simulation studies and applied to VAERS; implemented in R.