Identifying multifunctional therapeutic peptides (MFTP) is an important yet complex challenge in the realm of peptide recognition. Unlike monofunctional peptides, MFTP classification requires discerning fine-grained labeling information associated with amino acids, making it more intricate. Existing methods often ignore the nuanced semantics of these labels and fail to fully explore the interplay between peptide sequences and their labels. To address these issues, we propose a multilabel classification method named MultiPep-DLCL. This method uses a deep learning-based model architecture to translate peptide sequences into sequence features by learning the local and global dependencies of multifunctional therapeutic peptide sequences. Additionally, the Label-Sequence Fusion Transformer is employed to efficiently learn high-quality label embeddings by mining effective information from peptide sequences. Finally, the correspondence between sequence features and label embeddings is strengthened through label-sequence contrastive learning. To tackle dataset imbalance, MultiPep-DLCL integrates a multilabel focal dice loss function alongside the traditional cross-entropy loss function. Experimental results demonstrate that the MultiPep-DLCL significantly outperforms existing methods in MFTP recognition.

MOTIVATION: With the great number of peptide sequences produced in the postgenomic era, it is highly desirable to identify the various functions of therapeutic peptides quickly. Furthermore, it is a great challenge to predict accurate multi-functional therapeutic peptides (MFTP) via sequence-based computational tools. RESULTS: Here, we propose a novel multi-label-based method, named ETFC, to predict 21 categories of therapeutic peptides. The method utilizes a deep learning-based model architecture, which consists of four blocks: embedding, text convolutional neural network, feed-forward network, and classification blocks. This method also adopts an imbalanced learning strategy with a novel multi-label focal dice loss function. multi-label focal dice loss is applied in the ETFC method to solve the inherent imbalance problem in the multi-label dataset and achieve competitive performance. The experimental results state that the ETFC method is significantly better than the existing methods for MFTP prediction. With the established framework, we use the teacher-student-based knowledge distillation to obtain the attention weight from the self-attention mechanism in the MFTP prediction and quantify their contributions toward each of the investigated activities. AVAILABILITY AND IMPLEMENTATION: The source code and dataset are available via: https://github.com/xialab-ahu/ETFC.