Constrained variant detection with SPaRC: Sparsity, parental relatedness, and coverage.

Area of Science:

• Genomics
• Bioinformatics

Background:

• Structural variants (SVs) like inversions, deletions, insertions, and translocations are key DNA rearrangements.
• Traditional SV detection relies on mapping individual genome sequences to a reference.
• Large-scale sequencing studies now involve hundreds or thousands of related individuals.

Purpose of the Study:

• To develop a novel computational method for enhancing the detection of structural variants (SVs).
• To improve SV identification accuracy in large cohorts with related individuals.

Main Methods:

• Developed SPaRC (Sparsity, Parent-child, Relatedness, Coverage), an algorithmic formulation for SV detection.
• Incorporated realistic criteria including SV sparsity, individual relatedness, and variable sequencing coverage.
• Simultaneously analyzed parent and child genomes to boost true SV signals.

Main Results:

• SPaRC effectively boosts true SV signals by leveraging familial relationships.
• The algorithm accounts for genomic complexities like variable sequencing coverage.
• Demonstrated improved ability to identify SVs in large, related cohorts.

Conclusions:

• Simultaneous analysis of parent and child genomes significantly enhances SV detection.
• SPaRC provides a robust framework for SV identification in large-scale genomic studies.
• The method addresses limitations of traditional SV detection in complex cohort data.