In this study, we newly constructed datasets for eight different cell types. Utilizing these data, we proposed an integrative machine-learning-based framework called Enhancer-IF for identifying cell-specific enhancers. Enhancer-IF comprehensively explores a wide range of heterogeneous features with five commonly used machine learning methods (random forest, extremely randomized tree, multilayer perceptron, support vector machine, and extreme gradient boosting). Specifically, these five classifiers were trained with seven encodings and obtained 35 baseline models. The output of these baseline models is integrated and again inputted to five classifiers to construct five meta-models. Finally, the integration of five meta-models through ensemble learning improves the model robustness. The proposed approach showed an excellent prediction performance compared to the baseline models on both training and independent datasets in different cell types, highlighting the superiority of our approach in identifying enhancers. We assume that Enhancer-IF will be a valuable tool for screening and identifying potential enhancers from human DNA sequences.
