cWords

cWords identifies over-represented sequence motifs in differential case-control mRNA expression datasets to detect post-transcriptional regulatory elements, particularly small RNA (microRNA and siRNA) binding sites within 3' untranslated regions (UTRs).

Key Features:

• Motif Discovery: Uncovers sequence motifs within 3' untranslated regions (UTRs) of mRNAs, with explicit support for 6- and 7-nucleotide motifs indicative of microRNA binding sites.
• Performance Improvements: Algorithmic and statistical optimizations increase processing speed by at least 100-fold compared to previous versions.
• Benchmarking Success: In benchmark tests on 19 miRNA perturbation experiments, achieved equal or superior performance relative to miReduce and Sylamer.
• Versatility: Identifies potential siRNA off-target binding sites and endogenous microRNA binding motifs in mRNAs bound by Argonaute ribonucleoprotein particles.
• Statistical Rigor: Employs rigorous statistical methods to reduce bias and accurately identify over-represented motifs.
• Visualization and Clustering: Provides motif clustering and visualization capabilities to support analysis and interpretation of discovered motifs.

Scientific Applications:

• Post-transcriptional regulation analysis: Infers regulatory interactions mediated by small RNAs and RNA-binding proteins from motif enrichment in expression data.
• siRNA off-target detection: Detects potential off-target binding sites of siRNAs in transcriptome datasets.
• Argonaute-associated motif discovery: Discovers endogenous microRNA binding motifs within mRNAs associated with Argonaute ribonucleoprotein particles.
• Differential expression interpretation: Links over-represented sequence elements to changes observed in differential case-control mRNA expression experiments.
• Biological research contexts: Applicable to studies of developmental biology and disease pathology where post-transcriptional regulation is relevant.

Methodology:

Analyzes enrichment of 6- and 7-nucleotide motifs in 3' UTR sequences using statistical methods and optimized algorithms, followed by motif clustering and visualization, with performance validated by benchmarking against miReduce and Sylamer on 19 miRNA perturbation experiments.