MOTIVATION: Long non-coding RNAs (lncRNAs) have emerged as crucial players in diverse physiological and pathological processes, yet the biological mechanisms of the vast majority of lncRNAs remain elusive. To fill this gap, it is necessary to improve the accuracy of lncRNA identification and functional annotation. RESULTS: Here, we introduce LncADeep 2.0, an integrated deep learning framework designed to meet these needs. In the identification module, LncADeep 2.0 incorporated novel peptide features along with sequence and structural information, demonstrating superior performance over our previous LncADeep and other existing tools on both annotated transcripts from GENCODE and RNA-seq data. For functional annotation, LncADeep 2.0 leveraged lncRNA-centric interaction networks and gene ontology terms through the transfer learning strategy to achieve robust annotation performance with limited functional data. Compared to LncADeep, LncADeep 2.0 could accurately elucidate the general functions of given lncRNA sequences, predict tissue- or cell-type-specific functions from bulk and single-cell RNA-seq data, and establish connections between tumor-associated lncRNAs and genomic markers. Overall, LncADeep 2.0 stands out as an efficient and reliable tool for lncRNA identification and functional annotation across a wide spectrum of biological processes. AVAILABILITY AND IMPLEMENTATION: LncADeep 2.0 is available for use at https://github.com/Jefferson-Chou/LncADeep2 and https://doi.org/10.5281/zenodo.17164767.

Drug-drug interaction (DDI) refers to the interaction relationships between drugs. Discovering new DDIs is crucial for advancing drug development and enhancing clinical treatments. Given the significant progress achieved through graph neural networks (GNNs), network-based models have become a prevalent approach for tackling this challenge. However, current network-based approaches are incapable of seamlessly integrating a wide range of information. Motivated by this discovery, we propose a novel model, namely LLM-DDI, which aims to comprehensively tackle DDI prediction tasks by integrating various information of molecules in the BKG. LLM-DDI initially incorporates the generative pre-trained transformer (GPT) model to generate embeddings for each molecule within the biomedical knowledge graph (BKG). These embeddings encompass diverse types of information pertaining to each molecule. Subsequently, LLM-DDI utilizes a message-passing GNN framework to enhance the learning of molecular representations with the embeddings derived from GPT as input. LLM-DDI governs the propagation of information within the BKG by semantic relationships. These semantic relationships determine how information flows and is exchanged between different entities in the BKG. Finally, LLM-DDI leverages the learned drug representations to predict potential DDIs. Experiments show the effectiveness of LLM-DDI, as it achieves the best performance on two real-world datasets, providing valuable guidance for drug development and clinical treatment.

Accurately identifying cancer driver genes is crucial for understanding tumorigenesis and advancing precision oncology. However, integrating multi-omics data within complex biological networks remains challenging, particularly when it comes to capturing diverse structural information and leveraging the distinct signals from different omics modalities. While graphbased methods have demonstrated high accuracy in cancer gene identification, they might overlook the heterogeneity between omics features. To address this limitation, we propose CGI-MoE, a Mixture-of-Experts-inspired graph representation learning framework that incorporates omics-feature-specific expert modules based on graph transformers, together with adaptive gating and subgraph aggregation mechanisms. CGI-MoE extracts both local and global structural encodings for each node by sampling multiple subgraphs, enabling the model to capture comprehensive and robust network features. Each expert module focuses on a specific omics modality, and their outputs are fused by a lightweight gating network that dynamically weighs their contributions. When evaluated on both homogeneous and heterogeneous benchmark datasets, CGI-MoE achieves state-of-the-art performance in terms of accuracy, AUC, and AUPR, consistently surpassing existing methods. Ablation studies further highlight the critical roles of each component in achieving robust performance. Using the trained models, CGI-MoE predicted 46 novel cancer gene candidates from all unlabeled genes, demonstrating its potential for novel discovery and for deepening our understanding of cancer development. The code is available at https://github.com/moomight/CGI-MoE.