An improved molecular inversion probe based targeted sequencing approach for low variant allele frequency

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Deep targeted sequencing is crucial but limited by cost and complexity.
• Molecular Inversion Probes (MIP) offer scalability but face performance challenges, particularly in GC-rich regions and low variant allele frequencies.

Purpose of the Study:

• To enhance the performance and clinical applicability of MIP technology.
• To improve the detection of low-frequency variants using MIPs.
• To reduce the turnaround time and increase the efficiency of MIP-based sequencing.

Main Methods:

• Sequenced a large cohort (n=4417) using a 616-MIP panel at high depth.
• Developed and validated a statistical model for low variant allele frequency detection using machine learning.
• Implemented a biochemically modified MIP protocol and validated with amplicon sequencing.
• Assessed performance with a larger panel of 8349 genomic targets.

Main Results:

• Machine learning tools significantly improved precision for detecting variants with VAF > 0.005 (10-56.25%, P < 0.0004).
• The new MIP protocol improved uniformity, GC-rich region coverage, and achieved 95% on-target reads.
• Turnaround time was reduced to approximately 4 hours.
• Enhanced performance was demonstrated on a large panel of 8349 targets.

Conclusions:

• The study presents a significantly enhanced MIP targeted sequencing approach.
• Improvements in variant detection and overall performance pave the way for clinical diagnostics.
• The optimized MIP technology is positioned as an ultrafast, cost-effective tool for research and clinical applications.