metadata
license: mit
license: mit language: - en pretty_name: ABX-CT-003 Collateral Sensitivity Pattern tags: - antibiotics - combination-therapy - collateral-sensitivity - mic - resistance - tabular task_categories: - tabular-classification size_categories: - n<1k
ABX-CT-003 Collateral Sensitivity Pattern
Purpose
Detect the paired pattern where resistance to Drug A coincides with increased sensitivity to Drug B.
Core pattern
- stress_index high
- Drug A MIC crosses a_resistant_cutoff_mg_L
- Drug B MIC drops at least 2x vs baseline and reaches a low floor
- Drug B drop occurs at the same or next timepoint in v1
Files
- data/train.csv
- data/test.csv
- scorer.py
Schema
Each row is one timepoint in a within strain series.
Required columns
- row_id
- series_id
- timepoint_h
- organism
- strain_id
- drug_a
- drug_b
- stress_index
- baseline_mic_a_mg_L
- baseline_mic_b_mg_L
- mic_a_mg_L
- mic_b_mg_L
- a_resistant_cutoff_mg_L
- b_susceptible_floor_mg_L
- b_sensitivity_fold_change
- media
- assay_method
- source_type
- collateral_sensitivity_signal
- earliest_collateral_sensitivity
Labels
collateral_sensitivity_signal
- 1 for rows at or after the first detected collateral sensitivity row
earliest_collateral_sensitivity
- 1 only for the first detected row in that series
Scorer logic in v1
- baseline Drug A must be below resistant cutoff
- find first time Drug A crosses resistant cutoff under stress
- collateral sensitivity triggers if Drug B at the same or next timepoint
- drops by at least 2x vs baseline
- reaches b_susceptible_floor_mg_L or lower
Evaluation
Run
- python scorer.py --path data/test.csv