hic-compartment-shift

npx skills add majiayu000/claude-skill-registry --skill 32-hic-compartment-shift-bisnake2001-chromskills-2 --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

# Compartment shift Analysis --- ## Overview This skill performs A/B compartment shift analysis using PC1 eigenvector values extracted from Hi-C data, following the HOMER framework. It supports two conditions, each with two or more replicates, and uses the PC1 values (E1 column) from user-provided TSV files. Major steps include: - Refer to **Inputs & Outputs** to verify necessary files. - **Always prompt user** for genome assembly used. Never decide by yourself. - Convert TSV (Chrom, start, end, weight, E1) into HOMER-compatible PC1 bedGraph files. - Generate a unified genomic bin list for annotatePeaks. - Extract PC1 values across all samples. - Perform differential PC1 analysis with replicate-aware limma statistics. - Produce differential compartment tables and stitched compartment-shift domains. --- ## When to use this skill Use this skill when you want to: - Detect compartment shifts between two conditions (e.g., cell type 1 vs cell type 2) - Identify statistically significant changes in PC1 values across genomic bins - Determine regions that flip between A and B compartments - Integrate compartment shift results with other genomic datasets --- ## Inputs & Outputs ### Inputs Ex

compartments_shift_analysis/
shift_regions/
diff_PC1_CT2_vs_CT1.txt
temp/
bins_PC1.txt
PC1_all_samples.txt
awk 'BEGIN{OFS="	"} NR>1 && NF==5 {print $1, $2, $3, $5}' CT1_rep1.tsv > CT1_rep1.PC1.bedGraph
awk 'BEGIN{OFS="	"} NR>1 && NF==5 {print $1, $2, $3}' CT1_rep1.tsv > bins_PC1.txt