seqgendiff

seqgendiff simulates realistic RNA-seq datasets by adding controlled signals to existing count data to enable evaluation of RNA-seq analysis methods.

Key Features:

• Realistic data simulation: Adds known signals directly to existing RNA-seq count datasets to preserve complex and non-ideal characteristics of real data.
• Binomial thinning: Implements binomial thinning to introduce controlled biological and technical variation into counts.
• Function suite: R-based functions include select_counts() for selecting counts; thin_diff(), thin_lib(), thin_gene(), thin_2group(), and thin_all() for applying binomial thinning to simulate differential expression, library size variation, gene-specific changes, two-group comparisons, or comprehensive alterations; and effective_cor() to assess effective correlation.
• Applicability: Applicable to both single-cell and bulk RNA-seq datasets.

Scientific Applications:

• Method assessment and comparison: Generates realistic simulated datasets for rigorous assessment and comparison of RNA-seq analysis methods.
• Differential expression analysis: Provides simulated signals that can be used to validate and examine conclusions from differential expression studies.
• Factor analysis evaluation: Enables comparison of factor analysis techniques on RNA-seq datasets under realistic data conditions.

Methodology:

Adds known signals to real RNA-seq count data via binomial thinning and provides functions for selecting counts and assessing effective correlation (select_counts(), thin_diff(), thin_lib(), thin_gene(), thin_2group(), thin_all(), effective_cor()).