minnow

minnow simulates droplet-based single-cell RNA-sequencing (dscRNA-seq) experiments at the sequence level to produce realistic reads for validating analysis pipelines and assessing impacts on gene expression quantification.

Key Features:

• Sequence-Level Simulation: Simulates experiments at the sequence level, modeling polymerase chain reaction (PCR) amplification, cellular barcode (CB) and unique molecular identifier (UMI) selection, sequence fragmentation, and sequencing.
• Realistic Gene-Level Ambiguity: Replicates gene-level ambiguity observed in dscRNA-seq reads to enable evaluation of how ambiguous reads affect gene expression estimates.
• Impact Assessment of Processing Pipelines: Enables evaluation of read-alignment and UMI deduplication methods and their effects on gene-by-cell count matrices under realistic sequence ambiguity and experimental biases.

Scientific Applications:

• Pipeline validation: Validates read-alignment, UMI deduplication, and quantification pipelines using sequence-level simulated dscRNA-seq data.
• Method benchmarking: Assesses methods for handling ambiguous reads and quantifying their influence on gene expression measurements.
• Single-cell analyses: Supports pseudo-time series analysis, differential cell usage, cell-type detection, and RNA-velocity studies by providing controlled simulated datasets.

Methodology:

Simulates the full dscRNA-seq workflow from sample preparation through sequencing, explicitly modeling PCR amplification biases, cellular barcode (CB) and UMI selection, sequence fragmentation patterns, and sequencing.