Coverage clarification

[AmandaBirmingham]

Remove descriptions throughout of coverage as necessarily being a percentage, updated comments to indicate it can be either a percentage OR a fraction and that the user must verify that the min_coverage value is of the same type (percentage or fraction) as the coverage values. To make type mismatches less likely, remove the default value for the min_coverage parameter throughout so users must specify each time (hopefully in a way that matches the coverages they specify). Add explicit cast of coverage values to floating point in case coverage df is read in as strings.

[Copilot]

Pull request overview

This PR clarifies that OGU coverage values can be provided as either fractions or percentages, and makes corresponding updates to ensure consistency across documentation, code, and error messages. The key change is removing default values for min_coverage parameters to force users to explicitly specify thresholds that match their coverage value type (e.g., 0.01 for fractions vs 1 for percentages). The PR also adds explicit float casting for coverage values to handle cases where coverage data is read in as strings.

Key Changes

• Removed DEFAULT_MIN_PERCENT_COVERAGE constant and all default values for min_coverage parameters across public API functions
• Added OGU_AGNOSTIC_COVERAGE_KEY constant for internal use in long-format dataframes while maintaining OGU_PERCENT_COVERAGE_KEY for user-facing input
• Updated all documentation and error messages to refer to generic "coverage" instead of "% coverage" or "percent coverage" where appropriate
• Added explicit float casting in _generate_ogu_coverages_per_sample_df to handle string input

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
pysyndna/src/calc_cell_counts.py	Main changes including removal of DEFAULT_MIN_PERCENT_COVERAGE constant, addition of OGU_AGNOSTIC_COVERAGE_KEY for internal processing, parameter renaming from ogu_percent_coverage_df to ogu_coverage_df throughout, comprehensive documentation updates to clarify fraction vs percentage usage, removal of min_coverage defaults from all public functions, and addition of float casting for coverage columns
pysyndna/tests/test_calc_cell_counts.py	Test data structure updates to use OGU_AGNOSTIC_COVERAGE_KEY in expected output dictionaries, error message updates to remove "%" references, and addition of explicit string casting test for coverages_df to verify the new float casting functionality
pysyndna/src/fit_syndna_models.py	Minor comment improvement to be more accurate about checking multiple input dataframes

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.