HowDe-SBT

HowDe-SBT extends the Sequence Bloom Tree (SBT) framework to improve indexing and query efficiency for large-scale sequence collections such as RNA-seq datasets.

Key Features:

• Efficient Indexing: Implements a novel partitioning strategy that reduces index construction time and space requirements compared to traditional SBT methods.
• Performance Improvements: On real RNA-seq datasets, constructs indexes in less than 36% of the time required by previous SBT methods while using 39% less space.
• Fast Query Processing: Processes small-batch queries at least five times faster than existing SBT approaches.
• Theoretical Framework: Provides an analytical framework to bound and compare space and query performance relative to other SBT methods.

Scientific Applications:

• Transcriptomics (RNA-seq) indexing: Indexes and queries large transcript collections from RNA-seq datasets to support transcript-level analyses.
• Gene expression profiling: Enables rapid retrieval of transcripts for gene expression profiling.
• Differential expression analysis: Facilitates data access required for differential expression analysis across large sample sets.
• Large-scale genomic database querying: Supports efficient indexing and querying of large biological sequence databases for genomic investigations.

Methodology:

Builds on the SBT framework by implementing a partitioning mechanism for information storage and includes a theoretical framework to analyze and bound space and query performance.