RSSS

RSSS simulates and evaluates RNA sequencing (RNA-seq) experiments to compare transcript inference and quantification strategies under controlled conditions.

Key Features:

• Simulation framework: Simulates transcript sets, expression values, and sequencing reads across diverse parameter settings to model RNA-seq experiments under controlled conditions.
• Comparison of analysis strategies: Evaluates three principal approaches for transcript inference: curated annotations (assumes the sample transcriptome is a subset of the curated reference), genome-guided assembly (aligns reads to a reference genome and infers transcript structures, optionally leveraging curated annotations), and de novo assembly (reconstructs transcripts directly from reads without a reference genome or annotation set).
• Performance metrics: Quantifies sensitivity, precision, signal-to-noise characteristics, and identifies artefacts with high estimated expression that may propagate into downstream analyses.

Scientific Applications:

• Methodological assessment: Supports evidence-based selection or development of RNA-seq analysis strategies through empirical benchmarking.
• Transcriptome research: Enables investigation of how inference approaches behave under incomplete annotations and in the presence of artefact formation.
• Pipeline optimization: Helps tune analysis pipelines by selecting methods and parameterizations that balance sensitivity and precision for specific study goals.

Methodology:

Uses simulation-based study design to generate synthetic datasets with controlled ground truth across varying conditions, including incomplete annotation scenarios and artefact prevalence, and evaluates each method using sensitivity, precision, and related performance measures.