Robust Detection and Identification of Sparse Segments in Ultra-High Dimensional Data Analysis

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Copy number variants (CNVs) are significant genomic alterations involving deletions or duplications of DNA segments.
• CNVs range from kilobases to megabases and are implicated in various genetic conditions.
• Next-generation sequencing (NGS) data presents challenges for accurate CNV detection due to noise and segment sparsity.

Purpose of the Study:

• To develop a computationally efficient method for detecting and identifying sparse, short DNA segments indicative of CNVs.
• To provide a robust solution for segment identification that performs well across diverse noise distributions.
• To theoretically establish the conditions for signal detection and near-optimal estimation of CNV segments.

Main Methods:

• A novel computational method designed for identifying short, sparse segments within long linear data sequences.
• Robust statistical approach to handle unspecified noise distributions inherent in sequencing data.
• Theoretical analysis to quantify detection thresholds and estimation accuracy for CNV signals.

Main Results:

• The proposed method demonstrates computational efficiency and robustness across various noise models.
• Theoretical analysis confirms near-optimal signal segment estimation when detection is possible.
• Simulation studies validate the method's performance under different noise conditions.

Conclusions:

• The developed method offers a reliable and efficient tool for CNV detection using NGS data.
• The theoretical framework provides a strong basis for understanding the limits and capabilities of CNV detection algorithms.
• The approach is applicable to real-world genomic analyses, as shown by its use in a HapMap Yoruban sample CNV analysis.