simplifyEnrichment

simplifyEnrichment clusters similarity matrices of functional terms using the binary cut algorithm and provides functions for visualizing, summarizing, and comparing the resulting clusterings.

Key Features:

• Binary cut clustering: A new clustering algorithm that clusters similarity matrices of functional terms into groups with consistent within-group similarities and mutually exclusive between-group relationships.
• Visualization functions: Provides functions for visualizing the clustering results of functional enrichment terms.
• Summarization functions: Provides functions for summarizing the clustering results of functional enrichment terms.
• Comparison functions: Provides functions for comparing different clusterings of functional enrichment terms.

Scientific Applications:

• Functional enrichment result simplification: Reduces redundancy in long lists of significant functional terms from gene set enrichment analysis by clustering them into non-redundant, mutually exclusive groups.
• Benchmarking of similarity measures: Enables comparison of clustering performance across different similarity measures, including semantic similarity and gene overlap-based similarity matrices.

Methodology:

The binary cut algorithm clusters similarity matrices of functional terms. Benchmarks on simulated and real-world datasets compared binary cut performance across different similarity measures, including semantic similarity and gene overlap-based similarities.