ClarusC64/ABX-CT-001_synergy_score_decay-v0.1
ABX-CT-001 Synergy Score Decay Purpose Detect loss of combination synergy before monotherapy resistance appears. Core pattern stress_index high synergy_gap rises and stays high mono MICs stay below cutoffs during the early window later at least one mono MIC crosses its cutoff 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 synergy_score… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/ABX-CT-001_synergy_score_decay-v0.1.
ABX-CT-001 Synergy Score Decay
Purpose
Detect loss of combination synergy before monotherapy resistance appears.
Core pattern
- stress_index high
- synergy_gap rises and stays high
- mono MICs stay below cutoffs during the early window
- later at least one mono MIC crosses its cutoff
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
- synergy_score
- expectedsynergyscore
- synergy_gap
- monoAmicmg_L
- monoBmicmg_L
- monoAresistantcutoffmgL
- monoBresistantcutoffmgL
- media
- assay_method
- source_type
- synergydecaysignal
- earliestsynergydecay
Labels
- synergydecaysignal
- 1 for rows at or after the first confirmed synergy decay point
- earliestsynergydecay
- 1 only for the first detected decay row in that series
Scorer logic in v1
- baseline is timepoint 0
- candidate decay point
- stress_index at least 0.80
- synergy_gap at least 0.30 at two consecutive timepoints
- mono MICs below cutoffs and no more than 2x baseline
- exclude synergy_gap drop spike then snapback artifacts
- confirmation
- later monoA or monoB MIC crosses its cutoff
Evaluation
Run
- python scorer.py --path data/test.csv
