24,719
Unique complexes
Assembly–ligand pairs
DockTData
A receptor–ligand binding database connecting structure and bioactivity for discovery.
A FAIR-compliant dataset of receptor–ligand binding complexes for computational modeling and drug discovery.
Dataset at a glance
11,742
Distinct ligands
Unique InChIKeys
271
Protein–peptide complexes
Peptide ligand class
70
Protein–oligosaccharide complexes
Oligosaccharide ligand class
01 Dataset qualities
Curated for modern molecular research.
DockTData brings structural and energetic records into one consistent, research-ready foundation. Explore the dataset.
02 Method
Bridging structure and bioactivity.
DockTData is an automated extract, transform, and load pipeline designed to connect structural biology with molecular energetics.
Experimental structures from the Protein Data Bank are synchronized with bioactivity data from BindingDB and ChEMBL. The resulting receptor–ligand records are prepared for scalability, reproducibility, and use across machine-learning and molecular-modeling studies.
DockTData is developed by the same GMMSB group at LNCC behind the DockThor platform. Its machine-learning-ready records can also serve as a resource for training structure-prediction models such as AlphaFold and Boltz.
Dataset archive
2026-07-0 · 11.6 GB
Archive path
/downloads/snapshot_v2026-07-0/snapshot_v2026-07-0.zip
SHA-256
License
CC BY-SA 4.0
a07f46dd144dd845ac44d9f408053a902a5ae26349b585e7fe060fd77d80452b
03 Research team
Scientific infrastructure, built in Brazil.
The Grupo de Modelagem Molecular de Sistemas Biológicos at Laboratório Nacional de Computação Científica develops computational tools for molecular modeling and drug discovery research.
Visit GMMSB
GMMSB
The molecular-modeling research group behind DockTData.
LNCC
Based within Brazil's National Laboratory for Scientific Computing.
04 Open data
Bring DockTData into your research.
Start with the dataset or contact the team to discuss its preparation and use.