DLFF-ACP: prediction of ACPs based on deep learning and multi-view features fusion

Abstract

An emerging type of therapeutic agent, anticancer peptides (ACPs), has attracted attention because of its lower risk of toxic side effects. However process of identifying ACPs using experimental methods is both time-consuming and laborious. In this study, we developed a new and efficient algorithm that predicts ACPs by fusing multi-view features based on dual-channel deep neural network ensemble model. In the model, one channel used the convolutional neural network CNN to automatically extract the potential spatial features of a sequence. Another channel was used to process and extract more effective features from handcrafted features. Additionally, an effective feature fusion method was explored for the mutual fusion of different features. Finally, we adopted the neural network to predict ACPs based on the fusion features. The performance comparisons across the single and fusion features showed that the fusion of multi-view features could effectively improve the model’s predictive ability. Among these, the fusion of the features extracted by the CNN and composition of k-spaced amino acid group pairs achieved the best performance. To further validate the performance of our model, we compared it with other existing methods using two independent test sets. The results showed that our model’s area under curve was 0.90, which was higher than that of the other existing methods on the first test set and higher than most of the other existing methods on the second test set. The source code and datasets are available at https://github.com/wame-ng/DLFF-ACP.

Advancing Peptide-Based Cancer Therapy with AI: In-Depth Analysis of State-of-the-Art AI Models

“…cACP-DeepGram represented the peptides through fast-text embedding (FTE) through the skip-gram method (4-g), which captures the sequence order information in terms of 100 dimension features, and trained on dense neural network, which improved the prediction of anticancer peptides . A parallel neural network is stacked using a combination of dense network and CNN with a CNN-based feature extractor and compositional features (AAC, DPC, and composition of k-spaced amino acid group) in DLFF-ACP . The CNN-based AI4ACP method characterizes features using PC6 encoding schemes .…”

ACP-BC: A Model for Accurate Identification of Anticancer Peptides Based on Fusion Features of Bidirectional Long Short-Term Memory and Chemically Derived Information

“…This dataset contains a substantial number of non-ACP compounds without anticancer activity, simulating real-world positive-to-negative sample ratios. Additionally, we compared our ACP-BC model with three machine learning models AntiCP [22], iACP [30], ACPred-FL [38] and two deep learning models (DLFF-ACP [44], DeepACP [52]. Detailed data is presented in Table 4, revealing that the ACP-BC model's performance stands out with an ACC of 0.91, MCC of 0.40, specificity of 0.91, and an AUC of 0.92.…”

“…Yi et al [43] proposed ACP-DL, which selected BPF, a reduced amino acid alphabet, and the k-mer sparse matrix as features, and applied long short-term memory (LSTM) models for ACPs prediction. Cao et al [44] presented the DLFF-ACP model, using AAC, DPC, k-spaced amino acid group pairs (CKSAAGP), and Geary as features, and integrating deep learning and multi-view feature fusion for ACPs identification. Ahmed et al [40] recently developed APC-MHCNN, a computational model for predicting anticancer peptides that utilizes a multi-headed deep CNN.…”

Recent Progress in the Discovery and Design of Antimicrobial Peptides Using Traditional Machine Learning and Deep Learning

“…Regarding works on specific activity predictions for AMP, such as anticancer, Cao et al [ 135 ] proposed DLFF-ACP, a DL method using the DNN and CNN for predicting probabilities of ACPs. DLFF-ACP contains two input channels (also called the branch).…”