ClarusC64/ABX-CT-007_tissue_penetration_discordance-v0.1

ABX-CT-007 Tissue Penetration Discordance

Purpose

Detect when two drugs no longer reach the infection site together despite adequate plasma exposure.

Core pattern

• —stress_index high
• —siteefficacygap rises and stays high
• —sitecoexposureindex low
• —plasmaconcamgL and plasmaconcbmgL stay above simple floors
• —mono MICs stay below cutoffs at onset
• —latersitefailure_flag appears later

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
• —infection_site
• —organism
• —strain_id
• —drug_a
• —drug_b
• —stress_index
• —plasmaconcamgL
• —plasmaconcbmgL
• —siteconcamgL
• —siteconcbmgL
• —siteplasmaratio_a
• —siteplasmaratio_b
• —sitecoexposureindex
• —combositeefficacy_score
• —expectedcombositeefficacyscore
• —siteefficacygap
• —monoAmicmg_L
• —monoBmicmg_L
• —monoAresistantcutoffmgL
• —monoBresistantcutoffmgL
• —latersitefailure_flag
• —assay_method
• —source_type
• —tissuediscordancesignal
• —earliesttissuediscordance

Labels

• —tissuediscordancesignal
• —1 for rows at or after first confirmed discordance point

• —earliesttissuediscordance
• —1 only for the first detected discordance row in that series

Scorer logic in v1

• —candidate discordance point
• —stress_index at least 0.80
• —siteefficacygap at least 0.30 at two consecutive timepoints
• —sitecoexposureindex at most 0.35
• —plasma concentrations above simple floors
• —mono MICs below cutoffs and no more than 2x baseline
• —exclude siteefficacygap spike then snapback artifacts
• —confirmation
• —latersitefailure_flag equals 1 later in series

Evaluation

Run

• —python scorer.py --path data/test.csv