Systematic comparison of somatic variant calling performance among different sequencing depth and mutation frequency.

Area of Science:

• Oncology
• Genomics
• Bioinformatics

Background:

• Targeted cancer therapies rely on mutation detection, with next-generation sequencing (NGS) crucial for clinical research.
• Improving mutation calling performance is vital, often addressed by increasing sequencing depth.

Purpose of the Study:

• To evaluate the impact of sequencing depth and mutation frequency on somatic mutation detection.
• To compare the performance and speed of Strelka2 and Mutect2 variant callers.

Main Methods:

• Assessed 30 combinations of sequencing depth and mutation frequency.
• Utilized Strelka2 and Mutect2 for somatic mutation detection.
• Analyzed precision rates and caller performance across varying conditions.

Main Results:

• Precision rates exceeded 95% in most samples.
• For mutation frequencies ≥20%, ≥200X sequencing depth is adequate.
• For mutation frequencies ≤10%, experimental improvements are recommended over increased depth.
• Strelka2 outperformed Mutect2 at high frequencies; Mutect2 was better at low frequencies.
• Strelka2 demonstrated significantly faster processing speeds (17-22x).

Conclusions:

• Provides guidelines for optimizing sequencing depth and tool selection in clinical genomic research.
• Highlights the trade-offs between sequencing depth, mutation frequency, and caller performance.
• Recommends Strelka2 for speed and high-frequency mutation detection, and Mutect2 for low-frequency scenarios.