Biobambam2

Biobambam2 processes Next-Generation Sequencing (NGS) alignment files to perform name-based sorting, duplicate marking, and BAM-to-FASTQ conversion using an efficient collation algorithm.

Key Features:

• Fast sorting by read name: Efficiently sorts alignments by read name without requiring extensive memory, enabling processing of large BAM files.
• Duplicate marking: Marks duplicates within BAM files to support accurate downstream analyses, with improved efficiency on large and complex datasets.
• BAM-to-FASTQ conversion: Converts BAM files to FASTQ format rapidly, reported to outperform Picard and bamUtil in speed.
• Collation algorithm: Implements an algorithm that groups alignments by read name to minimize time and space consumption compared to full sorting.
• libmaus API exposure: Exposes the collation algorithm via the libmaus package for integration into other projects and workflows.

Scientific Applications:

• NGS preprocessing: Early-stage processing of alignment files for workflows such as variant calling, expression analysis, and structural variant detection.
• Read deduplication: Removal or marking of PCR/optical duplicates to improve data accuracy in whole-genome, whole-exome, and RNA-seq analyses.
• Re-alignment and reprocessing: Extraction of reads to FASTQ for re-alignment or for use with alternative alignment and analysis pipelines.
• Large-scale dataset handling: Processing of large sequencing datasets with reduced memory and runtime requirements.
• Pipeline integration: Incorporation of the collation algorithm into automated workflows via the libmaus API.

Methodology:

Uses an efficient collation algorithm that groups alignments by read name without full sorting, minimizing time and memory; the algorithm is exposed via the libmaus API.