BAMQL: a query language for extracting reads from BAM files.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomic Data Analysis

Background:

• Selecting subsets of reads from BAM files based on specific properties is a common bioinformatics task.
• Current methods using SAMtools and AWK are tedious, error-prone, and involve significant boilerplate code.
• Existing libraries like Bio-SamTools and pysam require substantial code for ad hoc filtering.

Purpose of the Study:

• To develop a more efficient and user-friendly method for filtering and extracting reads from BAM files.
• To provide a powerful query language for ad hoc analyses and integration into larger applications.

Main Methods:

• Development of a novel query language (BAMQL) for gathering reads based on predicates and logical connectives.
• Implementation of a system for compiling queries to native code for enhanced performance.

Main Results:

• BAMQL queries demonstrate faster execution compared to traditional equivalents.
• The query language supports predicates beyond those available in SAMtools, offering greater flexibility.
• Queries can be compiled to native code for efficient integration into larger programs.

Conclusions:

• BAMQL offers a user-friendly, powerful, and performant solution for extracting subsets of BAM files.
• It streamlines ad hoc analyses and simplifies integration into bioinformatics applications.
• The language's extended predicates and flexible connectives enhance its utility over existing tools.