GLAD

GLAD detects gains and losses in chromosomal regions from array Comparative Genomic Hybridization (array CGH) data to identify genomic alterations relevant to tumor progression and cancer biology.

Key Features:

• Breakpoint Detection: Employs the Adaptive Weights Smoothing (AWS) procedure to detect breakpoints that delineate altered genomic regions within array CGH profiles.
• Status Assignment: Assigns a status—gain, normal, or loss—to each chromosomal region based on detected breakpoints.
• Performance and Accuracy: Reports breakpoint detection rates of 97% in simulated data, 100% in karyotyping results, and 94% in manually analyzed profiles, with status assignment accuracy ranging from 98.9% to 99.8% for simulated data and 100% for karyotyping results.
• Benchmarking: Outperforms other existing solutions when evaluated against a public reference dataset.

Scientific Applications:

• Cancer Research: Identifies systematically deleted or amplified genomic regions to help pinpoint tumor suppressor genes in lost regions and oncogenes in gained regions.
• Tumor Classification: Generates detailed genomic profiles that can be used to refine tumor classification based on copy-number alterations.

Methodology:

Uses the Adaptive Weights Smoothing (AWS) procedure for breakpoint detection combined with a systematic status assignment that labels regions as gain, normal, or loss in array CGH profiles.