Detection Copy Number Variants from NGS with Sparse and Smooth Constraints

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Copy number variations (CNVs) are linked to complex diseases and cancers.
• Next-generation sequencing (NGS) offers rich data for CNV identification, but current methods lack consistent performance.
• Efficient and reliable CNV detection in NGS data remains a significant challenge.

Purpose of the Study:

• To develop a novel and efficient method for accurate copy number variation identification from next-generation sequencing data.
• To address the inconsistencies in performance among existing CNV detection tools.
• To improve the reconstruction of read depth signals for better CNV pattern fitting.

Main Methods:

• Formulated CNV identification as a constrained quadratic optimization problem.
• Incorporated sparsity and smoothness constraints to enhance signal reconstruction.
• Developed an efficient numerical solution using the Alternating Direction Minimization (ADM) framework, termed ADM-CNV.

Main Results:

• ADM-CNV successfully reconstructs CNV patterns from raw NGS data.
• The method demonstrates superior or comparable performance against six popular CNV detection approaches.
• Evaluated using synthetic, simulated, and empirical sequencing datasets, confirming robustness.

Conclusions:

• ADM-CNV offers a robust and accurate solution for copy number variation detection in next-generation sequencing data.
• The proposed optimization framework effectively leverages sparsity and smoothness for improved CNV identification.
• This method advances the reliable and efficient analysis of genomic variations in complex diseases and tumor types.