EAGLE: An algorithm that utilizes a small number of genomic features to predict tissue/cell type-specific enhancer-gene interactions
Enhancers were obtained by integrating diverse high-throughput datasets and the expressed levels were estimated using RNA-seq data. We utilized six features based on the information of enhancers and gene expression. ChIA-PET or Hi-C datasets were used to define positive and negative EG pairs. Using the labeled pairs, we trained an ensemble classifier, EAGLE, which could predict enhancer-target interactions measured by prediction probabilities.