DLFF-ACP

Anticancer peptides (ACPs), as a new type of therapeutic agent, have attracted more and more attention since they have lower toxic and side effects. However, it is a time-consuming and la-borious process to identify ACPs by experimental methods. Therefore, it is necessary to develop an efficient and accurate method for predicting ACPs. Here, we developed a new algorithm to predict ACPs by fusing different features based on deep learning. In the model, the convolutional neural network was used to extract the potential spatial features of the sequence automatically. In order to make full use of the physical and chemical properties of the peptide sequence, the handcrafted features were added to the inputs of the model. To realize the effective prediction of the ACPs, we fused different features and adopted the neural network to predict anticancer peptides. The single feature and fusion feature comparative experiments showed that the fusion of multiple features could effectively improve the predictive ability of the model. To further validate the performance of our model, we compared it with other existing methods on an independent test set, the results showed that the AUC of our model reached 88.7%, higher than other existing methods.