Skip to content

Examples

This page provides comprehensive examples of how to use cscore with simulations and real datasets.

Simulated scenarios

→ Open Interactive Notebook

The simulation demonstrates three fundamental scenarios: 1. Independent responses: No relationship 2. Common responses: Genes behave similarly in both comparisons 3. Divergent responses: Genes behave oppositely

Real Data Examples

The following examples use datasets provided in the testdata/ directory.

  1. GSE237099 - Bulk RNA-seq data (unloading/reloading muscle study)
  2. GSE236519 - Single-cell RNA-seq data (neuronal cell types)

Bulk RNA-seq Example (GSE237099)

This example demonstrates C-score analysis using bulk RNA-seq data from a muscle unloading/reloading study.

cscore -i testdata/GSE237099 \
  -a GSE237099_1_unloading_reloading_Reloading_vs_Control_deseq2.txt \
  -b GSE237099_1_unloading_reloading_Unloading_vs_Control_deseq2.txt \
  -o output_bulk_example.tsv \
  -n ensembl_gene_id \
  -e log2FoldChange \
  -f padj \
  -w 16 ## parallel workers

Expected Output:

The output file will contain columns including: - score: C-score (positive = common direction; negative = divergent) - p: permutation p-value for commonness/divergence - q_value: Benjamini–Hochberg adjusted p-value - convergence: "high" when commonness dominates, "low" otherwise - All original columns from both input files with _comp1/_comp2 suffixes

3. Single-cell RNA-seq Examples (GSE236519)

This dataset contains differential expression results for different neuronal cell types comparing knockout (KO) vs. knockdown (KD) conditions.

cscore -i testdata/GSE236519 \
  -a KO_Deep_Layer_neurons.txt \
  -b KD_Deep_Layer_neurons.txt \
  -o output_deep_layer_neurons.tsv \
  -n gene \
  -e logFC \
  -f FDR
  -w 16

Output Interpretation

C-score Values

  • Positive scores: Genes show consistent direction of change (both up or both down)
  • Negative scores: Genes show divergent responses (up in one condition, down in another)
  • Magnitude: Larger absolute values indicate stronger commonness/divergence

Significance Testing

  • p-value: Permutation-based significance of the observed C-score
  • q_value: Multiple testing corrected p-value (Benjamini-Hochberg)
  • convergence: Classification as "high" (common) or "low" (divergent) commonness

Getting Help

For additional options and help:

cscore -h