Dataset Viewer
Auto-converted to Parquet Duplicate
row_id
string
series_id
string
timepoint_h
int64
organism
string
strain_id
string
antibiotic_name
string
antibiotic_class
string
drug_trough_mg_L
float64
mic_mg_L
float64
media
string
assay_method
string
source_type
string
resistance_signal
int64
earliest_signal
int64
notes
string
ABXPD001-TR-0001
S1
0
Escherichia coli
EC-ATCC25922
meropenem
carbapenem
1.2
0.03
CAMHB
broth_microdilution
simulated
0
0
baseline stable
ABXPD001-TR-0002
S1
24
Escherichia coli
EC-ATCC25922
meropenem
carbapenem
2
0.03
CAMHB
broth_microdilution
simulated
0
0
trough up MIC stable
ABXPD001-TR-0003
S1
48
Escherichia coli
EC-ATCC25922
meropenem
carbapenem
2.1
0.03
CAMHB
broth_microdilution
simulated
0
0
stable
ABXPD001-TR-0004
S2
0
Klebsiella pneumoniae
KP-CLIN001
ceftriaxone
3rd_gen_cephalosporin
18
0.5
CAMHB
broth_microdilution
simulated
0
0
baseline
ABXPD001-TR-0005
S2
24
Klebsiella pneumoniae
KP-CLIN001
ceftriaxone
3rd_gen_cephalosporin
20
0.5
CAMHB
broth_microdilution
simulated
0
0
stable coverage
ABXPD001-TR-0006
S2
48
Klebsiella pneumoniae
KP-CLIN001
ceftriaxone
3rd_gen_cephalosporin
19
2
CAMHB
broth_microdilution
simulated
1
1
first MIC rise breaks coverage
ABXPD001-TR-0007
S2
72
Klebsiella pneumoniae
KP-CLIN001
ceftriaxone
3rd_gen_cephalosporin
22
4
CAMHB
broth_microdilution
simulated
1
0
continued resistance
ABXPD001-TR-0008
S3
0
Staphylococcus aureus
SA-ATCC29213
vancomycin
glycopeptide
12
1
CAMHB
broth_microdilution
simulated
0
0
baseline
ABXPD001-TR-0009
S3
24
Staphylococcus aureus
SA-ATCC29213
vancomycin
glycopeptide
14
16
CAMHB
broth_microdilution
simulated
0
0
unit or read error spike
ABXPD001-TR-0010
S3
48
Staphylococcus aureus
SA-ATCC29213
vancomycin
glycopeptide
13
1
CAMHB
broth_microdilution
simulated
0
0
snap back supports error

ABX-PD-001: MIC Coherence Decay

This dataset tests early resistance detection.

You detect the first time MIC stops tracking drug exposure.

Core idea

You track

  • drug trough concentration in mg/L
  • MIC in mg/L
  • ratio = trough / MIC

You flag resistance when

  • MIC rises fast
  • trough stays stable or rises
  • ratio drops hard

You avoid false alarms from

  • PK failure where trough collapses
  • measurement spikes that snap back

Files

  • data/train.csv
  • data/test.csv
  • scorer.py

Schema

Each row is one timepoint in a longitudinal series.

Required columns

  • row_id
  • series_id
  • timepoint_h
  • organism
  • strain_id
  • antibiotic_name
  • antibiotic_class
  • drug_trough_mg_L
  • mic_mg_L
  • media
  • assay_method
  • source_type
  • resistance_signal
  • earliest_signal

Labels

  • resistance_signal

    • 1 for rows at or after emergence
    • 0 otherwise
  • earliest_signal

    • 1 only for the first detected row in that series
    • 0 otherwise

Evaluation

Run

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

The scorer reports

  • F1 for resistance_signal
  • earliest signal hits and misses
  • PK failure flags
  • measurement spike flags

Notes on realism

This v0.1 seed includes

  • stable controls
  • clear emergence
  • slow drift emergence
  • trough drop cases
  • spike and snap cases
Downloads last month
20