enrichment

enrichment performs enrichment analysis to identify enriched sequence motifs and gene set associations from ranked lists of DNA sequences for gene set enrichment and sequence element discovery.

Key Features:

• Statistical Framework: Implements principled partitioning of data into target and background sets, rigorous models for measuring motif enrichment including exact p-values, an explicit framework to account for motif multiplicity, and controls to reduce false positives from randomly generated data.
• Software Application (DRIM): Embodied in DRIM (Discovery of Rank Imbalanced Motifs), which identifies sequence motifs within ranked lists of DNA sequences and is applicable to datasets such as ChIP-chip and CpG methylation.

Scientific Applications:

• ChIP-Chip Data: In yeast ChIP-chip analysis DRIM identified 50 novel putative transcription factor binding sites, elucidated aspects of the transcriptional network of TF ARO80, and revealed systematic TF binding enhancements to sequences containing CA repeats.
• CpG Methylation Data: In human cancer CpG methylation data DRIM discovered novel motifs similar to DNA sequence elements bound by the Polycomb complex, supporting a mechanistic link between histone methylation and CpG methylation.

Methodology:

DRIM analyzes ranked lists of DNA sequences to discover enriched sequence motifs using principled target/background partitioning, statistical models that yield exact p-values, mechanisms to account for motif multiplicity, and procedures to minimize false positives.