hi-c-coverage-track-computation

hi-c-coverage-track-computation

Summary

Compute per-bin sequencing depth (coverage) from cooler-formatted Hi-C contact matrices using cooltools.coverage(), producing a normalized track of bin-level read abundance across the genome. This quantifies the aggregate sequencing depth at each genomic bin, a key quality control and normalization step in high-resolution Hi-C analysis.

When to use

Apply this skill when you have a cooler file (.cool or .mcool) from a Hi-C experiment and need to generate a genome-wide track of per-bin sequencing depth to assess coverage uniformity, identify poorly sequenced regions, or normalize downstream analyses by local sequencing intensity. Use it before downstream Hi-C computations (e.g., insulation, contact scaling, saddle point analysis) that may be confounded by uneven sequencing depth.

When NOT to use

• Input is already a pre-computed coverage track or bigWig file — skip directly to downstream use.
• Analysis goal requires only the total contact count per chromosome, not per-bin resolution.
• Cooler file is empty or contains no valid contact pairs (malformed or failed sequencing).

Inputs

• cooler file (.cool or .mcool format) — a HDF5-based contact matrix from Hi-C sequencing
• cooler.Cooler object loaded in memory from a cooler file

Outputs

• pandas Series indexed by genomic bins with per-bin coverage values (sequencing depth)
• bedGraph, CSV, or TSV file with columns: chromosome, start, end, coverage

How to apply

Install cooltools in development mode (pip install -e .) to access bundled test datasets and the cooltools library. Load a cooler object from a .cool or .mcool file using the cooler library (e.g., cooler.Cooler(path)). Call cooltools.coverage() on the loaded cooler object, optionally specifying whether to store total cis counts back into the cooler file metadata. The function returns a pandas Series or track indexed by genomic bins with coverage values (total sequencing depth per bin). Export the result to a tabular format (bedGraph, CSV, or TSV) containing bin coordinates (chrom, start, end) and corresponding coverage values. Validate the output by checking row counts match the total number of bins in the cooler file and that coverage values are non-negative and non-zero for expected genomic regions.