Transformer-CNN: Fast and Reliable tool for QSAR
This work addresses the need for fast and reliable tools in cheminformatics for drug discovery and chemical property prediction, offering an incremental improvement with enhanced interpretability and small dataset handling.
The authors tackled the problem of improving QSAR/QSPR model quality and interpretability by developing Transformer-CNN, a method that uses SMILES embeddings from a Transformer and CharNN architecture, achieving higher performance on diverse benchmark datasets for regression and classification tasks.
We present SMILES-embeddings derived from the internal encoder state of a Transformer [1] model trained to canonize SMILES as a Seq2Seq problem. Using a CharNN [2] architecture upon the embeddings results in higher quality interpretable QSAR/QSPR models on diverse benchmark datasets including regression and classification tasks. The proposed Transformer-CNN method uses SMILES augmentation for training and inference, and thus the prognosis is based on an internal consensus. That both the augmentation and transfer learning are based on embeddings allows the method to provide good results for small datasets. We discuss the reasons for such effectiveness and draft future directions for the development of the method. The source code and the embeddings needed to train a QSAR model are available on https://github.com/bigchem/transformer-cnn. The repository also has a standalone program for QSAR prognosis which calculates individual atoms contributions, thus interpreting the model's result. OCHEM [3] environment (https://ochem.eu) hosts the on-line implementation of the method proposed.