PXR Challenge Train/Test Dataset

A high-quality experimental dataset for predicting human Pregnane-X Receptor (PXR) induction, comprising over 11,000 compounds screened using a high-fidelity in-house assay. This is the largest publicly available PXR activity dataset, released as part of the OpenADMET PXR Induction Blind Challenge.

Blog post: Announcing the Next OpenADMET Blind Challenge: Predicting PXR Induction

Challenge Space: openadmet/pxr-challenge

Challenge period: April 1 – July 1, 2026

Produced by: OpenADMET

CHANGELOG

• Updated 2026-09-28 Added raw dose-response curve (DRC) well-level data as downloadable parquet files (pxr-challenge_DRC_plot_data_TRAIN.parquet / pxr-challenge_DRC_plot_data_TEST_UNBLINDED.parquet, with real unblinded curves for the train/test set). Only available via pd.read_parquet (see below).
• Updated 2026-09-02 Corrected SMILES for 10 compounds in pxr-challenge_96-compound-uscale-semi-pure_TRAIN.csv (OCNT-2469084, -2469095, -2469107, -2469108 through -2469114). Regiochemical enumeration errors; all corrections are constitutional isomers (ΔMW = 0) so no activity or yield values changed. Re-pull if you have trained on a previous version of semi_pure_htchem.
• Updated 2026-07-03 Added phase 2 unblinded labels for the default test set subset in phase_2_unblinded config.
• Updated 2026-05-27 Added phase 1 unblinded labels for the default test set subset in phase_1_unblinded config.
• Updated 2026-05-27 Additional crude data added, see here for more detail.
• Updated 2026-04-09 dropping some compounds, fixing minor confidence interval issues and improving naming join. See here for more details.

Dataset contents

Two additional downloadable parquet files — pxr-challenge_DRC_plot_data_TRAIN.parquet and pxr-challenge_DRC_plot_data_TEST_UNBLINDED.parquet, with the raw per-well datapoints reconstructed from the assay's well-level data, i.e. the actual dose-response curves that pEC50/Emax in default were fit from.

Loading with Hugging Face datasets

from datasets import load_dataset

# Default config (primary assay)
ds = load_dataset("openadmet/pxr-challenge-train-test")
train = ds["train"]
test  = ds["test"]

# Counter-assay config
ds_counter = load_dataset("openadmet/pxr-challenge-train-test", "counter_assay")
train_counter = ds_counter["train"]

# Structure config
ds_structure = load_dataset("openadmet/pxr-challenge-train-test", "structure")
test_structure = ds_structure["test"]

# Single-concentration config
ds_single = load_dataset("openadmet/pxr-challenge-train-test", "single_concentration")
train_single = ds_single["train"]

# crudes htchem config
ds_crudes = load_dataset("openadmet/pxr-challenge-train-test", "crudes_htchem")
train_crudes = ds_crudes["train"]

# semi-pures htchem config
ds_semi_pure = load_dataset("openadmet/pxr-challenge-train-test", "semi_pure_htchem")
train_semi_pure = ds_semi_pure["train"]

# Phase 1 unblinded test labels config
ds_phase1 = load_dataset("openadmet/pxr-challenge-train-test", "phase_1_unblinded")
test_phase1 = ds_phase1["test"]

# Phase 2 unblinded test labels config
ds_phase2 = load_dataset("openadmet/pxr-challenge-train-test", "phase_2_unblinded")
test_phase2 = ds_phase2["test"]

Loading directly with pandas

import pandas as pd

train          = pd.read_csv("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_TRAIN.csv")
test           = pd.read_csv("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_TEST_BLINDED.csv")
train_counter  = pd.read_csv("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_counter-assay_TRAIN.csv")
test_structure = pd.read_csv("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_structure_TEST_BLINDED.csv")
train_single   = pd.read_csv("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_single_concentration_TRAIN.csv")
train_crudes   = pd.read_csv("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_htchem-libraries_TRAIN.csv")
train_semi_pure = pd.read_csv("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_96-compound-uscale-semi-pure_TRAIN.csv")
test_phase1    = pd.read_csv("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_TEST_PHASE_1_UNBLINDED.csv")
test_phase2    = pd.read_csv("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_TEST_PHASE_2_UNBLINDED.csv")
train_drc      = pd.read_parquet("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_DRC_plot_data_TRAIN.parquet")
test_drc       = pd.read_parquet("hf://datasets/openadmet/pxr-challenge-train-test/pxr-challenge_DRC_plot_data_TEST_UNBLINDED.parquet")