B3DB: Blood-Brain Barrier Permeability Database, 7,807 Compounds with BBB+/BBB- Labels and 1,058 logBB Values
- Provider
- Fanwang Meng, Paul W. Ayers et al. (theochem); mirror by Mao Lab, University of Michigan
- License
- CC0-1.0 (source)
- Size
- 10 files
- Format
- CSV (in sampled files)
- Stored at
- Hugging Face (https://huggingface.co/datasets/maomlab/B3DB)
- Access
- Free and public. No account with the data provider is needed.
- Category
- Drug Discovery
- Source
- https://www.nature.com/articles/s41597-021-01069-5
About this dataset
B3DB, the Blood-Brain Barrier Database, is a benchmark of small-molecule blood-brain barrier (BBB) permeability compiled from 50 published sources and categorized by the agreement between experimental references. It labels 7,807 compounds as BBB-permeable (BBB+) or not (BBB-), and gives numerical logBB values, the log ratio of brain to blood concentration, for 1,058 of them, together with computed physicochemical descriptors. This Hugging Face copy, prepared by the Mao Lab at the University of Michigan, mirrors the theochem/B3DB GitHub repository (the 2021 classification and 2023 regression releases) as train and test CSV configs: B3DB_classification, B3DB_classification_extended and B3DB_regression, with invalid SMILES removed or replaced in version 1.2.0 and a raw data summary of the sources. Use it to train and benchmark CNS drug-likeness and BBB penetration models for neuroscience drug discovery; load with datasets or pandas. License CC0 1.0, from the B3DB repository. Cite Meng et al., A curated diverse molecular database of blood-brain barrier permeability with chemical descriptors, Scientific Data 8, 289 (2021).
How to load this dataset in Python
import pandas as pd
df = pd.read_csv("hf://datasets/maomlab/B3DB/B3DB_classification/test.csv")
df.head()Data structure
File types in sampled files: CSV
Files
Example file: B3DB_classification/test.csv
Analyze this dataset without downloading it
On Scigantic, B3DB: Blood-Brain Barrier Permeability Database, 7,807 Compounds with BBB+/BBB- Labels and 1,058 logBB Values opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.
This page is also available as Markdown and as JSON.
Related Drug Discovery datasets
- GDB Databases
- Modelling of ready biodegradability based on combined public and industrial data sources
- OpenADMET AHR Induction: Aryl Hydrocarbon Receptor Reporter Assay Screen of 11,360 Compounds with Dose-Response Follow-Up
- Alchemistry 2027 Antiviral Blind Challenge: Test Set
- SaLT&PepPr: An Interface-Predicting Language Model for Designing Peptide-Guided Protein Degraders
- SurGBSA: Molecular Dynamics Trajectories and MM/GBSA Scores for PDBbind CASF-2016
- GWAS pyrethroid-resistant Aedes aegypti
- PubChem CID-SMILES topology classification snapshot