SLAM-DUNK

SLAM-DUNK quantifies T > C nucleotide conversions arising from 4-thiouridine labeling in RNA sequencing (SLAMseq) and related metabolic RNA-labeling protocols to measure labeled transcript dynamics.

Key Features:

• Digital Unmasking (DUNK) integration: Leverages the Digital Unmasking of Nucleotide conversions in K-mers (DUNK) pipeline for conversion-aware analysis.
• Constant mapping rates: Maintains consistent mapping rates irrespective of nucleotide-conversion frequencies.
• Recovery of multimapping reads: Recovers and accounts for reads that map to multiple genomic locations.
• SNP masking: Employs Single Nucleotide Polymorphism (SNP) masking to distinguish true SNPs from nucleotide conversions.
• Normalization approaches: Provides raw counts of conversion-containing reads and normalized estimates based on base content and read coverage to estimate labeled transcript fractions.
• Validation: Validated using experimentally generated and simulated datasets to assess robustness and sensitivity in quantifying conversions.

Scientific Applications:

• SLAMseq data analysis: Quantifies T > C conversions in SLAMseq datasets to detect and measure 4-thiouridine-labeled transcripts.
• Time-resolved RNA dynamics: Estimates labeled transcript fractions and RNA turnover dynamics over time from metabolic labeling experiments.
• Support for related protocols: Applies to other time-resolved RNA-sequencing methods such as TimeLapse-seq and TUC-seq.
• SNP versus conversion discrimination: Differentiates nucleotide-conversion signals from genetic variation to reduce false positives.

Methodology:

Uses the DUNK pipeline; performs conversion-aware mapping that maintains constant mapping rates, recovers multimapping reads, applies SNP masking, and outputs raw counts plus normalized estimates based on base content and read coverage; validation employed experimental and simulated datasets.