yara

yara aligns DNA sequencing reads to reference genomes using approximate seeds and multiple backtracking to accelerate approximate search and improve alignment specificity and sensitivity.

Key Features:

• Approximate seeds for filtration: Uses approximate seeds to enhance filtration specificity without compromising sensitivity compared to exact-seed methods.
• Multiple backtracking: Applies multiple backtracking to efficiently manage large sets of seeds by leveraging the repetitiveness of next-generation sequencing data.
• Accelerated approximate search: Reduces computational costs and accelerates approximate searches on genomic datasets through its filtration and backtracking strategies.
• Performance benchmarks: Reports performance improvements over RazerS 3 and mrFAST by an order of magnitude, and outperforms Bowtie 2 and BWA in both speed (2–4×) and accuracy.
• Implementation detail: Implemented in C++ using the SeqAn library.

Scientific Applications:

• Read mapping: Mapping DNA sequencing reads to reference genomes for genomic analyses.
• High-accuracy alignment: Producing precise alignments of next-generation sequencing reads by balancing sensitivity and specificity.
• Approximate-search on genomic datasets: Accelerating approximate-search operations on repetitive NGS datasets.

Methodology:

Employs approximate-seed filtration and multiple backtracking to manage large seed sets and speed up approximate searches, implemented in C++ with the SeqAn library.