pblat

pblat parallelizes BLAT to perform high-throughput nucleotide sequence alignment of DNA and RNA sequences against reference genomes and transcriptomes.

Key Features:

• Parallelization and Multithreading: Implements multithreaded processing to leverage multicore processors and reduce runtime for large-scale alignments.
• Cluster Computing Support: Provides a pblat-cluster variant that distributes tasks across multiple nodes using Message Passing Interface (MPI) for scalable performance.
• Memory Efficiency: Maintains memory usage comparable to BLAT while increasing throughput.
• Accuracy and Precision: Produces results identical to BLAT, preserving alignment precision and output format.

Scientific Applications:

• Long and gapped sequence alignment: Aligns long sequences and supports gapped mappings that are challenging for short-read mappers.
• Genomic and transcriptomic fine-mapping: Rapidly fine-maps large-scale DNA and RNA sequences against reference genomes.
• Genome-wide association studies: Supports high-throughput mapping requirements for GWAS.
• Comparative and functional genomics: Enables precise alignments for comparative genomics and functional genomics analyses.

Methodology:

Parallelizes the original BLAT algorithm via multithreading on multicore systems and an MPI-based pblat-cluster for distributed computing, while preserving BLAT's outputs and memory footprint.