HaarSeg

HaarSeg segments array Comparative Genomic Hybridization (aCGH) data to partition probes into regions with consistent DNA copy number for detection of genomic copy-number alterations.

Key Features:

• Wavelet-based segmentation: Uses Haar wavelet decomposition and thresholding, exploiting maxima of the Haar wavelet transform to locate breakpoints in aCGH signals.
• Breakpoint detection: Identifies statistically significant breakpoints to define segments of equal copy number.
• Computational efficiency: Operates over 1000 times faster than existing leading approaches as reported.
• Adaptability: Algorithmic structure allows generalization and adaptation to incorporate additional side information.
• Reliability weighting: Integrates measurement reliability indicators to weight probes during segmentation.
• Noise variability compensation: Compensates for changes in measurement noise variability to support robust segmentation across conditions.

Scientific Applications:

• Genome-wide CNV detection: Scanning the genome using aCGH data to detect DNA copy number variations.
• Cancer and genetic disorder studies: Identification of genetic abnormalities associated with cancer and other genetic disorders via segmented copy-number profiles.

Methodology:

Performs Haar wavelet decomposition and thresholding, uses maxima of the Haar wavelet transform to detect statistically significant breakpoints, and can integrate reliability indicators and compensate for measurement noise variability.