Multi-ACPNet Integrates Sequence-Structure Features For Anticancer Peptide Prediction
Lu Meng and Lijun Zhou publish Multi-ACPNet in PLoS Computational Biology on March 10, 2026, presenting a dual-function deep-learning framework that fuses sequence and structural features using a BiLSTM–causal convolution encoder and a multi-scale graph convolutional network. The model reports identification accuracies of 0.8140, 0.9536, and 0.8770 on three benchmark datasets and functional-prediction metrics of AUC 0.9033, F1 0.8472, and Hamming loss 0.1303, outperforming prior methods.
Scoring Rationale
Peer-reviewed methodological advance with usable code and webserver, but incremental novelty confined to peptide-discovery applications.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
