GREAT:SCAN:patterns

GREAT:SCAN:patterns detects periodic patterns in the positions of genomic features to characterize genome architecture, cluster in-phase features, map genomic regions with periodicities, and support transcription factor binding site prediction through integration with multi-view machine learning.

Key Features:

• Automated detection of periodic patterns: Systematically identifies periods from the positions of genomic features across the genome.
• Clustering of in-phase genomic features: Groups genomic features that align in phase for each detected period to reveal positional co-organization.
• Mapping of genomic regions with periodicities: Maps genome regions exhibiting significant periodic patterns to provide spatial context for regulatory elements.
• Multi-view machine learning integration: Integrates periodicity and positional information in a multi-view machine learning framework to enhance prediction of transcription factor binding sites.

Scientific Applications:

• Genome architecture analysis: Characterizes the spatial organization of co-functional genes and regulatory elements across chromosomes.
• Gene regulation and chromosomal organization studies: Links genomic layout with gene expression and chromosomal architecture.
• Transcription factor binding site prediction: Improves prediction of transcription factor binding sites by incorporating periodicity and positional features.
• Epigenetics and regulatory element investigation: Identifies positional periodicities associated with regulatory elements to support studies in gene regulation and epigenetics.

Methodology:

Detection: identifying all potential periodic patterns from the positions of genomic features; Clustering: grouping genomic features that align in-phase for each detected period; Mapping: localizing regions across the genome where these periodicities occur.