hot_scan

hot_scan identifies genomic regions unusually rich in specific patterns by using scan statistics to detect translocation breakpoints and other sequencing-based hotspots.

Key Features:

• Probabilistic model integration: Uses a probabilistic model within the scan statistics framework to determine both the number and location of genomic hotspots.
• Global chromosome-wide control level: Provides a global chromosome-wide nominal control level for clustering analysis rather than relying on local significance alone.
• Versatility across experimental approaches: Applicable to analyses motivated by chromosomal translocations in activated B lymphocytes and to data from techniques such as ChIP-seq and 4C-seq.
• Implementation: Implemented in R and Perl.

Scientific Applications:

• Translocation hotspot detection: Identifies regions frequently translocated to oncogenes such as c-myc and can define hotspots that are longer than those identified by previous methods.
• Impact of genetic variations: Detects changes in hotspot size associated with genetic perturbations, including absence of the DNA repair protein 53BP1 and combinations with overexpression of activation-induced cytidine deaminase.
• Identification of exclusive hotspots: Pinpoints exclusive translocation hotspots within genes known to be tumor suppressors.

Methodology:

hot_scan applies scan statistics integrated with a probabilistic model to analyze next-generation sequencing data for detection of unusual clusters and is implemented in R and Perl.