datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
TDC23-RedTeaming
TDC 2023 (LLM Edition) - Red Teaming Track
This is the combined dev and test set from the Red Teaming Track of TDC 2023.
Citation
If find this dataset useful, please cite the following work:
@inproceedings{tdc2023,
title={TDC 2023 (LLM Edition): The Trojan Detection Challenge},
author={Mantas Mazeika and Andy Zou and Norman Mu and Long Phan and Zifan Wang and Chunru Yu and Adam Khoja and Fengqing Jiang and Aidan O'Gara and Ellie Sakhaee and Zhen Xiang and Arezoo… See the full description on the dataset page: https://huggingface.co/datasets/walledai/TDC23-RedTeaming.TDC_herg_central_at_10umtest
TDC_pampa_ncats
TDC PAMPA NCATS
PAMPA NCATS dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
NCATS subset of PAMPA dataset.
PAMPA (parallel artificial membrane permeability assay) is an assay to evaluate drug permeability across the cellular membrane. The task models only the passive membrane diffusion. This is “NCATS” subset of the dataset created at National Center for Advancing Translational Sciences (NCATS).
This dataset is a part of… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_pampa_ncats.TDC_caco2_wang
TDC Caco-2 Wang
Caco-2 Wang dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict the rate at which drug passes through Caco-2 cells that serve as in vitro simulation of human intestinal tissue.
This dataset is a part of "absorption" subset of ADME tasks.
Characteristic
Description
Tasks
1
Task type
regression
Total samples
910
Recommended splitscaffold
Recommended metric
MAE… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_caco2_wang.TDC_herg_central_at_1umTDC_solubility_aqsoldb
TDC Solubility AqSolDB
Solubility AqSolDB dataset dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
he task is to predict the aqeuous solubility - a measure drug's ability to dissolve in water.
Poor water solubility could lead to slow drug absorptions,
inadequate bioavailablity and even induce toxicity.
This dataset is a part of "absorption" subset of ADME tasks.
Characteristic
Description
Tasks
1
Task type… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_solubility_aqsoldb.TDC_herg_central_inhibTDC_cyp2d6_veith
CYP P450 2D6 Inhibition, Veith et al.
CYP P450 2D6 Inhibition dataset by Veith et al., part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict the binding results for a set of inhibitors of CYP2D6 enzyme, part of cytochromes P450 family.
Characteristic
Description
Tasks
1
Task type
classification
Total samples
1513
Recommended split
scaffold
Recommended metric
AUROC
References
[1]
Veith… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_cyp2d6_veith.TDC_cyp1a2_veith
CYP P450 1A2 Inhibition, Veith et al.
CYP P450 1A2 Inhibition dataset by Veith et al., part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict the binding results for a set of inhibitors of CYP1A2 enzyme, part of cytochromes P450 family.
Characteristic
Description
Tasks
1
Task type
classification
Total samples
12579
Recommended split
scaffold
Recommended metric
AUROC
References
[1]
Veith… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_cyp1a2_veith.TDC_pgp_broccatelli
TDC P-gp Inhibition
P-gp (P-glycoprotein) Inhibition dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict whether a molecule inhibits the P-glycoprotein.
P-glycoprotein (P-gp) is an ABC transporter protein involved in intestinal absorption, drug metabolism,
and brain penetration, and its inhibition can seriously alter a drug's bioavailability and safety
Characteristic
Description
Tasks
1
Task type… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_pgp_broccatelli.TDC_bioavailability_ma
TDC Bioavailability
Bioavailability dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict the oral bioavailability.
It is defined as "the rate and extent to which the active ingredient or active moiety
is absorbed from a drug product and becomes available at the site of action".
This dataset is a part of "absorption" subset of ADME tasks.
Characteristic
Description
Tasks
1
Task type
classification… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_bioavailability_ma.TDC_cyp2c19_veith
CYP P450 2C19 Inhibition, Veith et al.
CYP P450 2C19 Inhibition dataset by Veith et al., part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict the binding results for a set of inhibitors of CYP2C19 enzyme, part of cytochromes P450 family.
Characteristic
Description
Tasks
1
Task type
classification
Total samples
12665
Recommended split
scaffold
Recommended metric
AUROC
References
[1]
Veith… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_cyp2c19_veith.TDC_cyp3a4_veith
CYP P450 3A4 Inhibition, Veith et al.
CYP P450 3A4 Inhibition dataset by Veith et al., part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict the binding results for a set of inhibitors of CYP3A4 enzyme, part of cytochromes P450 family.
Characteristic
Description
Tasks
1
Task type
classification
Total samples
12328
Recommended split
scaffold
Recommended metric
AUROC
References
[1]
Veith… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_cyp3a4_veith.TDC_ames
TDC AMES
AMES dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict mutagenicity of drugs, i.e. potential to induce
genetic alterations. This data comes from a standardized Ames test,
which is a short-term bacterial reverse mutation assay.
Characteristic
Description
Tasks
1
Task type
classification
Total samples
7278
Recommended splitscaffold
Recommended metric
AUROC
References… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_ames.TDC_cyp2c9_veith
CYP P450 2C9 Inhibition, Veith et al.
CYP P450 2C9 Inhibition dataset by Veith et al., part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict the binding results for a set of inhibitors of CYP2C9 enzyme, part of cytochromes P450 family.
Characteristic
Description
Tasks
1
Task type
classification
Total samples
12092
Recommended split
scaffold
Recommended metric
AUROC
References
[1]
Veith… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_cyp2c9_veith.TDC_vdss_lombardo
TDC VDss
Volume of Distribution at Steady State dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict the volume of distribution at steady state (VDss) that measures
the degree of drug's concentration in body tissue compared to concentration in blood [1]_ [2]_.
This dataset is a part of "distribution" subset of ADME tasks.
Characteristic
Description
Tasks
1
Task type
classification
Total… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_vdss_lombardo.TDC_hia_hou
TDC HIA HOU
Human Intestinal Absorption dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict whether a drug is well absorbed via the human intestine. It is relevant for oral drug design.
This dataset is a part of "absorption" subset of ADME tasks.
Characteristic
Description
Tasks
1
Task type
classification
Total samples
578
Recommended splitscaffold
Recommended metric
AUROC… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_hia_hou.TDC-DrugADMETTDC_ld50_zhu
TDC Acute Toxicity LD50
Acute Toxicity LD50 dataset dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
Acute toxicity LD50 measures the most conservative dose that
can lead to lethal adverse effects.
The regression task is to predict the acute toxicity of drugs.
Characteristic
Description
Tasks
1
Task type
regression
Total samples
7385
Recommended split
scaffold
Recommended metric
MAE
References… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_ld50_zhu.TDC_herg_karim
TDC Herg Karim
Herg Karim dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict whether the drug blocks ether-à-go-go related gene (hERG),
crucial for coordination of the heart's beating.
This dataset is a binary classification task, with molecules defined as hERG (<10 µM)
and non-hERG (>=10 µM) blockers.
Characteristic
Description
Tasks
1
Task type
classification
Total samples
13445… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_herg_karim.TDC_carcinogens_lagunin
TDC Carcinogens Lagunin
Carcinogens Lagunin dataset dataset [1] [2], part of TDC [3] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict whether the drug is a carcinogen.
Characteristic
Description
Tasks
1
Task type
classification
Total samples
280
Recommended split
scaffold
Recommended metricAUROC
References
[1]
Lagunin, Alexey, et al.
Computer-Aided Prediction of Rodent Carcinogenicity by PASS and… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_carcinogens_lagunin.TDC
Therapeutics Data Commons
Therapeutics Data Commons(TDC) provides a publicly available collection of 22 machine learning tasks for therapeutic discovery. Our Hugging Face repository is a mirror of single-instance prediction tasks of TDC, encompassing a total of 46,265,659 data points.
The parquet files uploaded to our Hugging Face repository have been sanitized and curated from the original datasets.
Each parquet file corresponds to a separate category (e.g., ADME) and contains… See the full description on the dataset page: https://huggingface.co/datasets/maomlab/TDC.TDC_skin_reaction
TDC Skin Reaction
Skin Reaction dataset dataset [1], part of TDC [2] benchmark. It is intended to be used through
scikit-fingerprints library.
The task is to predict whether the drug can cause immune reaction that leads to skin sensitization.
Characteristic
Description
Tasks
1
Task type
classification
Total samples
404
Recommended split
scaffold
Recommended metricAUROC
References
[1]
Alves, Vinicius M., et al.
"Predicting chemically-induced skin… See the full description on the dataset page: https://huggingface.co/datasets/scikit-fingerprints/TDC_skin_reaction.TDC_clearance_microsome_azTDC_pampa_approved_drugsTDC_ppbr_azTDC_hergTDC_half_life_obachTDC_diliTDC_rlm
