Introduction to Deep-B³

A deep learning based multi-model framework model to predict the BBB permeability of candidate compounds. the samples are encoded in three kinds of features, namely molecular descriptors and fingerprints, images, and simplified molecular input line entry system (SMILES) text notation. The pre-trained models were built to extract latent features from images and SMILES. These features depicted the compounds in terms of tabular data, image and text, respectively. The validation results yielded from the independent dataset demonstrated that the performance of Deep-B3 is superior to that of the state-of-the-art models.