Fatemeh Nazem’s research while affiliated with Isfahan University of Medical Sciences and other places

One of the critical aspects of structure-based drug design is to choose important druggable binding sites in the protein’s crystallography structures. As experimental processes are costly and time-consuming, computational drug design using machine learning algorithms is recommended. Over recent years, deep learning methods have been utilized in a wide variety of research applications such as binding site prediction. In this study, a new combination of attention blocks in the 3D U-Net model based on semantic segmentation methods is used to improve localization of pocket prediction. The attention blocks are tuned to find which point and channel of features should be emphasized along spatial and channel axes. Our model’s performance is evaluated through extensive experiments on several datasets from different sources, and the results are compared to the most recent deep learning-based models. The results indicate the proposed attention model can predict binding sites accurately, i.e. the overlap of the predicted pocket using the proposed method with the true binding site shows statistically significant improvement when compared to other state-of-the-art models. The attention blocks may help the model focus on the target structure by suppressing features in irrelevant regions.

Background: The first step in developing new drugs is to find binding sites for a protein structure that can be used as a starting point to design new antagonists and inhibitors. The methods relying on convolutional neural network for the prediction of binding sites have attracted much attention. This study focuses on the use of optimized neural network for three-dimensional (3D) non-Euclidean data. Methods: A graph, which is made from 3D protein structure, is fed to the proposed GU-Net model based on graph convolutional operation. The features of each atom are considered as attributes of each node. The results of the proposed GU-Net are compared with a classifier based on random forest (RF). A new data exhibition is used as the input of RF classifier. Results: The performance of our model is also examined through extensive experiments on various datasets from other sources. GU-Net could predict the more number of pockets with accurate shape than RF. Conclusions: This study will enable future works on a better modeling of protein structures that will enhance knowledge of proteomics and offer deeper insight into drug design process.

Binding site prediction for new proteins is important in structure-based drug design. The identified binding sites may be helpful in the development of treatments for new viral outbreaks in the world when there is no information available about their pockets with Covid-19 being a case in point. Identification of the pockets using computational methods, as an alternative method, has recently attracted much interest. In this study, the binding site prediction is viewed as a semantic segmentation problem. An improved 3D version of the U-Net model based on the dice loss function is utilized to predict the binding sites accurately. The performance of the proposed model on the independent test datasets and SARS-COV-2 shows the segmentation model could predict the binding sites with a more accurate shape than the recently published deep learning model, i.e. DeepSite. Therefore, the model may help predict the binding sites of proteins and could be used in drug design for novel proteins.