Fig. 1
From: XGBCDA: a multiple heterogeneous networks-based method for predicting circRNA-disease associations
The Flowchart of the XGBCDA method. The XGBCDA method includes the following steps: extract statistical features and graph theory features from an integrated circRNA similarity network, an integrated disease similarity network and circRNA-disease association network; input these original features into the XGBoost classifier to further obtain latent features; integrate the latent features with the original features to train the final XGBoost classifier for predicting circRNA-disease association