BLSSPELLER

BLSSPELLER identifies conserved cis-regulatory motifs in orthologous promoter sequences across related species by enumerating IUPAC-word motifs and scoring their conservation for comparative genomics analyses.

Key Features:

• Alignment-free and alignment-based discovery: implements an algorithm that enumerates putative motifs as words over the IUPAC alphabet and screens them for conservation using a branch length score, enabling motif discovery without reliance on pre-generated multiple sequence alignments.
• Comprehensive scoring system: assigns a genome-wide confidence score to identified motifs to distinguish functionally relevant signals from background.
• Cloud computing (MapReduce): leverages the MapReduce programming model to enable scalable, parallelized processing of large genomic datasets.
• Enrichment analysis: reports enrichment of high-scoring motifs in open chromatin regions in Oryza sativa and in transcription factor binding sites inferred from protein-binding microarrays in Oryza sativa and Zea mays.
• Experimental validation recovery: recovers experimentally profiled ga2ox1-like KN1 binding sites in Zea mays.

Scientific Applications:

• Comparative motif discovery in plant genomics: identification of conserved cis-regulatory elements across related species to support studies of gene regulation and evolutionary conservation.
• Integration with chromatin and PBM data: prioritization and validation of predicted motifs using open chromatin maps and protein-binding microarray–inferred transcription factor binding sites in Oryza sativa and Zea mays.
• Regulatory network inference: support for delineating regulatory networks related to plant development and stress responses through conserved motif identification.

Methodology:

Exhaustive enumeration of potential motifs within orthologous promoter sequences as IUPAC words, conservation screening using a branch length score, computation of genome-wide motif confidence scores, support for both alignment-free and alignment-based discovery, and scalable computation via the MapReduce model.