Blood Brain Barrier Permeablility Predictor
Conventional experimental approaches for BBB permeability measurement are expensive, time-consuming, and labor-intensive. Thus, accurate prediction of blood brain barrier (BBB) permeability of a compound is a major challenge in neurotherapeutic drug discovery. We developed a BBB permeability prediction model (LightBBB) using a large dataset for practical applications. This model can be used for facilitated compound screening in early stage of brain drug discovery.

Server Submission

Since LightBBB calculates molecular descriptors relies on SMILES (Simplified Molecular-Input Line-Entry System), jobs can currently only be submitted by using SMILES.

Please enter correct SMILES. To predict toxicity of multiple molecules simultaneously, SMILES file with name extension '.smi' can alternatively be queried. For this query file, please input SMILES line by line.


or Submit .SMI File:

Citation: Shaker, B., Yu, M., Song, J.S. et al. LightBBB: Computational prediction model of blood-brain-barrier penetration based on LightGBM. Bioinformatics btaa918,20 (2020)

Bug reports: blisszen [at]

Dataset used for BBB model is available at:

School of integrative engineering
Chung-Ang University, Korea