Multi-ACPNet: A multi-scale sequence-structure feature fusion framework for anticancer peptide identification and functional prediction

Anticancer peptides (ACPs) have emerged as promising therapeutic candidates for cancer treatment due to their high efficacy and low propensity for inducing drug resistance. However, existing ACP identification methods primarily rely on peptide sequence features while neglecting spatial structural characteristics. Moreover, few approaches can simultaneously predict the functional activity of ACPs. To address these limitations, this study proposes Multi-ACPNet, a novel dual-function predictor capable of both ACP identification and activity type classification. This model innovatively integrates sequence and structural features through a multi-stage framework. It employs a hybrid Bidirectional Long Short-Term Memory (BiLSTM) and causal convolutional network to capture both long-range dependencies and local sequence patterns, followed by a multi-scale Graph Convolutional Network (GCN) that dynamically fuses local and long-range structural dependencies using residual connections and adaptive weighting. Experimental results demonstrate that Multi-ACPNet achieves outstanding performance, with Accuracy of 0.8140, 0.9536, and 0.8770 on three benchmark datasets for ACP identification. For functional prediction, it attains an AUC of 0.9033, F1-score of 0.8472, and Hamming loss of 0.1303, significantly outperforming state-of-the-art predictors.Author summary: Anticancer peptides (ACPs) have emerged as highly promising therapeutic candidates in cancer treatment due to their high efficacy and low propensity for drug resistance. However, existing prediction methods suffer from two major limitations: first, they predominantly rely on sequence information while neglecting three-dimensional structural features, leading to the loss of crucial spatial interaction information; second, their predictive capability is insufficient—most current models are limited to binary classification tasks for ACP identification and cannot predict their targeting activities against specific cancer cell lines. To address these challenges, we developed Multi-ACPNet, an end-to-end framework that integrates both sequence and structural information for dual-task collaborative prediction. The model innovatively incorporates a multi-scale information fusion mechanism: at the sequence level, it combines Bidirectional Long Short-Term Memory (BiLSTM) with causal convolution to simultaneously capture long-range dependencies and local functional motifs; at the structural level, a multi-hop Graph Convolutional Network is constructed to dynamically learn multi-level spatial interactions within peptide molecules. This architecture not only significantly improves the accuracy of ACP identification but also enables efficient multi-label prediction of functional activities across seven cancer cell types. Furthermore, the model demonstrates excellent generalization capability, achieving competitive performance in extended tasks such as peptide toxicity prediction.