datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
selfies-ids-cleanedbelka-selfies-idsbelka-selfies-train-cls-ftsmiles-selfies-pretrainPubChem-124M-SMILES-SELFIES-InChI-IUPAC
PubChem-124M-Canonicalized-SELFIES-InChI-IUPAC
Dataset Summary
This dataset contains ~124 million chemical structures sourced from PubChem (as of Jan 2026), processed into a clean, machine-learning-ready Parquet format.
Unlike raw XML/JSON dumps or standard CSVs, this dataset provides a unified, tabular structure that joins multiple chemical identifiers and descriptors into a single sharded resource:
SMILES: Raw and RDKit-Canonicalized.
SELFIES: Pre-computed… See the full description on the dataset page: https://huggingface.co/datasets/Bilsteen/PubChem-124M-SMILES-SELFIES-InChI-IUPAC.selfies-train-idspubchem-selfies-pretrainchemnlp-selfies-pretrainbelka-selfies-train-cls-ft-idsZINC-selfies-20mPubChem-SMILES-SELFIES-InChI-IUPAC-v2enamine-nature-selfies-pretrain-p1PubChem10M_SMILES_SELFIESpubchem-10m-selfiesChEMBL36-SELFIES
ChEMBL 36 SELFIES
Pre-training dataset for ModernMolBERT. Contains ~2.4M drug-like small molecules from ChEMBL 36 represented as SELFIES strings.
Dataset details
field
value
source
lukaskim/ChEMBL-36
representation
SELFIES
train rows
2,390,314
validation rows
24,228
total rows
2,414,542
min heavy atoms
3
max heavy atoms
100
max MW
1000.0
deduplicated by
InChIKey
split method
deterministic hash on InChIKey
valid fraction
0.01… See the full description on the dataset page: https://huggingface.co/datasets/HauserGroup/ChEMBL36-SELFIES.PubChem-SMILES-SELFIES-InChI-IUPAC-v3enamine-diverse-selfies-pretrainftinder_selfies
Dataset Card for "ftinder_selfies"
More Information needed
molenet-selfies-pretrainPLINDER-selfiesguacamol-selfies-pretrainGLASS_GPCR_SELFIES
From the GLASS GPCR database, converted to SELFIES
Steps to prepare the database:
Download the GLASS database
wget https://zhanggroup.org/GLASS/downloads/interactions_active.tsv
wget https://zhanggroup.org/GLASS/downloads/interactions_inactives.tsv
wget https://zhanggroup.org/GLASS/downloads/targets.tsv
wget https://zhanggroup.org/GLASS/downloads/ligands.tsv
Select just the columns of interest
cut -d$'\t' -f6,9 ligands.tsv > ligands2.tsv
cut -d$'\t' -f2,5 targets.tsv >… See the full description on the dataset page: https://huggingface.co/datasets/introvoyz041/GLASS_GPCR_SELFIES.belka-selfies-testLD50_selfies_logLD50_selfiesPubChem10M_Canonicalized_SELFIES
Dataset Card for "PubChem10M_Canonicalized_SELFIES"
More Information needed
PLINDER-selfiesZINC_4M_SELFIES
From ZINC20 'In-stock, Lead-like' tranche, converted to SELFIES
Steps to prepare the database:
Select the appropriate tranche from from ZINC20
Select 'Purch' -> 'In-stock'
Select 'Predefined Subsets' -> 'Lead-Like'
Select 'Download Format' -> 'SMILES (*.smi)'
Select 'Download Method' -> 'Raw URLs'
Download and concatenate the SMILES
# Download all ZINC20 tranches from 'in-stock, lead-like' subset
mkdir zinc
wget -i ZINC-downloader-2D-smi.uri -P zinc
# Remove first line of… See the full description on the dataset page: https://huggingface.co/datasets/introvoyz041/ZINC_4M_SELFIES.enamine-diverse-selfies-pretrainmoses_selfies_cleaned