MPscan

MPscan performs index-free exact read mapping for next-generation sequencing (NGS) datasets by executing exact pattern matching in a single pass over the reference genome to enable efficient high-throughput exact matching for transcriptomic and epigenomic analyses.

Key Features:

• High-throughput exact mapping: Implements exact read mapping suitable for large-scale NGS datasets and millions of short reads.
• Index-free single-pass matching: Performs exact pattern matching in a single pass over the reference genome without constructing an index, distinguishing it from FM-index/seed-and-extend approaches.
• Filtration scheme: Employs a filtration-based pruning strategy that reduces the search space and enables sublinear average-time complexity with respect to genome length.
• Concurrent processing: Processes millions of short reads concurrently during a single genome traversal to avoid repeated genome scans.
• Low memory footprint and no preprocessing: Maintains minimal memory usage and avoids costly genome preprocessing or index construction.
• Benchmark performance: Demonstrates runtime performance improvements over BLAT-like aligners and several specialized mappers such as Bowtie and Zoom for exact matching workloads.
• Scalability and accuracy evaluations: Accuracy and scalability assessments indicate index-free exact matching remains viable for specific read mapping applications.

Scientific Applications:

• Exact matching workloads: Suited for tasks that require exact pattern matches of reads to a reference genome.
• Transcriptomic analyses: Applicable to transcriptomic pipelines that rely on exact matching subroutines.
• Epigenomic analyses: Applicable to epigenomic pipelines that perform exact read mapping.
• Algorithm development and benchmarking: Serves as a reference method for benchmarking and for informing optimized approximate mapping strategies that build on exact matching.

Methodology:

Performs exact pattern matching in a single pass over the reference genome without constructing an index and employs a filtration scheme that prunes the search space to achieve sublinear average-time complexity while processing millions of short reads concurrently.