Prediction of Estrogen Receptor alpha Antagonists Using Deep Neural Network.

Breast cancer is one of the most common diseases whose seriousness has raised concern and anxiety worldwide with over 7 million cases and around 685 thousand deaths globally (according to the World Health Organization(WHO)), making it one of the world’s most prevailing diseases. Scientists and doctors tried to cure breast cancer disease using many types of therapeutic treatment, the most common of them being Endo-therapy. Unfortunately however, Endo-therapy did ot have much better results than the other types of treatments. Some breast cancer cells have estrogen receptors which can attract estrogen proteins and make a cancer cell grow. In this study, we introduce deep neural networks on QSAR model to find inhibitors for those estrogen receptors and thus stop the breast cancer cells from growing. We present a comprehensive comparison between our model (the DNNR model) and the Random forest model, in which the DNNR outperformed the RF algorithm (an increase of 21.2% in the accuracy of the algorithm). [ABSTRACT FROM AUTHOR]