DeePred-BBB is a deep Neural network-based model for prediction of blood brain barrier permeability of compounds using Simplified Molecular Input Line Entry System (SMILES) notation of Compounds.

The files contained in this repository are as follows:

• DeePred-BBB_Script.py: main script to run predictions
• smiles.smi: user input structures (multiple)
• DeePredmodel.h5: DNN prediction model
• data.csv: training dataset
• PaDEL: folder with executable for feature calculation

• Python
• Numpy
• Pandas
• Keras
• Tensorflow

In order to run BBB permeability predictions with DeePred-BBB, save input structures as SMILES in a single file (e.g. smiles.smi). Remember to give a name or ID to each structure.

1. Download this repository and ensure that all the files are present in the same folder when running the script.
2. Run DeePred-BBB_Scrip.py.

python DeePred-BBB_Script.py <folder>

If <folder> is not provided, the script runs in the current directory. A csv file (DeePred-BBB_predictions.csv) will be created in the folder where the script is run. A file containing the features generated by PaDEL will be also saved to disk (PaDEL_features.csv).

NOTE: Remember to activate the corresponding conda environment before running the script, if applicable.

If you use DeepPred-BBB in your publication, consider citing the paper:

@ARTICLE{10.3389/fnins.2022.858126,
AUTHOR={Kumar, Rajnish and Sharma, Anju and Alexiou, Athanasios and Bilgrami, Anwar L. and Kamal, Mohammad Amjad and Ashraf, Ghulam Md},   
TITLE={DeePred-BBB: A Blood Brain Barrier Permeability Prediction Model With Improved Accuracy},      
JOURNAL={Frontiers in Neuroscience},      
VOLUME={16},           
YEAR={2022},     
URL={https://www.frontiersin.org/articles/10.3389/fnins.2022.858126},       
DOI={10.3389/fnins.2022.858126},      	
ISSN={1662-453X}
}