MutScan: fast detection and visualization of target mutations by scanning FASTQ data.

Area of Science:

• Genomics
• Bioinformatics

Background:

• Clinical genetic tests, including circulating tumor DNA (ctDNA) analysis for cancer, necessitate sensitive detection of known target mutations.
• Conventional next-generation sequencing (NGS) pipelines often involve filtering steps that can lead to the omission of low-frequency mutations.
• Existing variant validation tools are often resource-intensive, making them impractical for ultra-deep sequencing data.

Purpose of the Study:

• To develop a tool, MutScan, for the sensitive detection and efficient validation of target mutations.
• To address the limitations of conventional NGS pipelines in identifying low-frequency mutations and validating variants.

Main Methods:

• MutScan utilizes highly optimized string-searching algorithms to scan FASTQ files for reads supporting target mutations.
• It employs rolling hash and bloom filter algorithms for accelerated scanning and detection.
• Supporting reads are piled up and visualized using web technologies (HTML, JavaScript) for interactive analysis.

Main Results:

• MutScan achieves high performance, operating approximately 20 times faster than conventional pipelines.
• It offers maximal sensitivity, capable of detecting mutations with as few as one supporting read.
• The tool generates interactive visualizations for mutation validation, aiding in the avoidance of false positives.

Conclusions:

• MutScan provides direct detection and visualization of target mutations from FASTQ raw data.
• It enhances performance and sensitivity compared to traditional methods, crucial for applications like ctDNA analysis.
• MutScan facilitates cloud-friendly VCF validation through HTML page generation and is available as an open-source tool.