Bacteriophages are viruses that target bacteria, playing a crucial role in microbial ecology. Phage proteins are important in understanding phage biology, such as virus infection, replication, and evolution. Although a large number of new phages have been identified via metagenomic sequencing, many of them have limited protein function annotation. Accurate function annotation of phage proteins presents several challenges, including their inherent diversity and the scarcity of annotated ones. Existing tools have yet to fully leverage the unique properties of phages in annotating protein functions. In this work, we propose a new protein function annotation tool for phages by leveraging the modular genomic structure of phage genomes. By employing embeddings from the latest protein foundation models and Transformer to capture contextual information between proteins in phage genomes, GOPhage surpasses state-of-the-art methods in annotating diverged proteins and proteins with uncommon functions by 6.78% and 13.05% improvement, respectively. GOPhage can annotate proteins lacking homology search results, which is critical for characterizing the rapidly accumulating phage genomes. We demonstrate the utility of GOPhage by identifying 688 potential holins in phages, which exhibit high structural conservation with known holins. The results show the potential of GOPhage to extend our understanding of newly discovered phages.

Accurate prediction of virus-host interactions is critical for understanding viral ecology and developing applications like phage therapy. However, the growing number of computational tools has created a complex landscape, making direct performance comparison challenging due to inconsistent benchmarks and varying usability. Here, we provide a systematic review and a rigorous benchmark of 27 virus-host prediction tools. We formulate the host prediction task into two primary frameworks-link prediction and multi-class classification-and construct two benchmark datasets to evaluate tool performance in distinct scenarios: a database-centric dataset (RefSeq-VHDB) and a metagenomic discovery dataset (MetaHiC-VHDB). Our results reveal that no single tool is universally optimal. Performance is highly context-dependent, with tools like CHERRY and iPHoP demonstrating robust, broad applicability, while others, such as RaFAH and PHIST, excel in specific contexts. We further identify a critical trade-off between predictive accuracy, prediction rate, and computational cost. This work serves as a practical guide for researchers and establishes a standardized benchmark to drive future innovation in deciphering complex virus-host interactions.

MOTIVATION: Nucleocytoplasmic large DNA viruses (NCLDVs) are notable for their large genomes and extensive gene repertoires, which contribute to their widespread environmental presence and critical roles in processes such as host metabolic reprogramming and nutrient cycling. Metagenomic sequencing has emerged as a powerful tool for uncovering novel NCLDVs in environmental samples. However, identifying NCLDV sequences in metagenomic data remains challenging due to their high genomic diversity, limited reference genomes, and shared regions with other microbes. Existing alignment-based and machine learning methods struggle with achieving optimal trade-offs between sensitivity and precision. RESULTS: In this work, we present GiantHunter, a reinforcement learning-based tool for identifying NCLDVs from metagenomic data. By employing a Monte Carlo tree search strategy, GiantHunter dynamically selects representative non-NCLDV sequences as the negative training data, enabling the model to establish a robust decision boundary. Benchmarking on rigorously designed experiments shows that GiantHunter achieves high precision while maintaining competitive sensitivity, improving the F1-score by 10% and reducing computational cost by 90% compared to the second-best method. To demonstrate its real-world utility, we applied GiantHunter to 60 metagenomic datasets collected from six cities along the Yangtze River, located both upstream and downstream of the Three Gorges Dam. The results reveal significant differences in NCLDV diversity correlated with proximity to the dam, likely influenced by reduced flow velocity caused by the dam. These findings highlight GiantHunter's potential to advance our understanding of NCLDVs and their ecological roles in diverse environments. AVAILABILITY AND IMPLEMENTATION: The source code of GiantHunter is available via: https://github.com/FuchuanQu/GiantHunter.

MOTIVATION: Plasmids play an essential role in horizontal gene transfer, aiding their host bacteria in acquiring beneficial traits like antibiotic and metal resistance. There exist some plasmids that can transfer, replicate, or persist in multiple organisms. Identifying the relatively complete host range of these plasmids provides insights into how plasmids promote bacterial evolution. To achieve this, we can apply multi-label learning models for plasmid host range prediction. However, there are no databases providing the detailed and complete host labels of these broad-host-range plasmids. Without adequate well-annotated training samples, learning models can fail to extract discriminative feature representations for plasmid host prediction. RESULTS: To address this problem, we propose a self-correction multi-label learning model called MOSTPLAS. We design a pseudo label learning algorithm and a self-correction asymmetric loss to facilitate the training of multi-label learning model with samples containing some unknown missing labels. We conducted a series of experiments on the NCBI RefSeq plasmid database, the PLSDB 2025 database, plasmids with experimentally determined host labels, the Hi-C dataset, and the DoriC dataset. The benchmark results against other plasmid host range prediction tools demonstrated that MOSTPLAS recognized more host labels while keeping a high precision. AVAILABILITY AND IMPLEMENTATION: MOSTPLAS is implemented with Python, which can be downloaded at https://github.com/wzou96/MOSTPLAS. All relevant data we used in the experiments can be found at https://zenodo.org/doi/10.5281/zenodo.14708999.

MOTIVATION: Viruses, with their ubiquitous presence and high diversity, play pivotal roles in ecological systems and public health. Accurate identification of viruses in various ecosystems is essential for comprehending their variety and assessing their ecological influence. Metagenomic sequencing has become a major strategy to survey the viruses in various ecosystems. However, accurate and comprehensive virus detection in metagenomic data remains difficult. Limited reference sequences prevent alignment-based methods from identifying novel viruses. Machine learning-based tools are more promising in novel virus detection but often miss short viral contigs, which are abundant in typical metagenomic data. The inconsistency in virus search results produced by available tools further highlights the urgent need for a more robust tool for virus identification. RESULTS: In this work, we develop ViraLM for identifying novel viral contigs in metagenomic data. By using the latest genome foundation model as the backbone and training on a rigorously constructed dataset, the model is able to distinguish viruses from other organisms based on the learned genomic characteristics. We thoroughly tested ViraLM on multiple datasets and the experimental results show that ViraLM outperforms available tools in different scenarios. In particular, ViraLM improves the F1-score on short contigs by 22%. AVAILABILITY AND IMPLEMENTATION: The source code of ViraLM is available via: https://github.com/ChengPENG-wolf/ViraLM.

MOTIVATION: Protein embedding, which represents proteins as numerical vectors, is a crucial step in various learning-based protein annotation/classification problems, including gene ontology prediction, protein-protein interaction prediction, and protein structure prediction. However, existing protein embedding methods are often computationally expensive due to their large number of parameters, which can reach millions or even billions. The growing availability of large-scale protein datasets and the need for efficient analysis tools have created a pressing demand for efficient protein embedding methods. RESULTS: We propose a novel protein embedding approach based on multi-teacher distillation learning, which leverages the knowledge of multiple pre-trained protein embedding models to learn a compact and informative representation of proteins. Our method achieves comparable performance to state-of-the-art methods while significantly reducing computational costs and resource requirements. Specifically, our approach reduces computational time by ∼70% and maintains ±1.5% accuracy as the original large models. This makes our method well-suited for large-scale protein analysis and enables the bioinformatics community to perform protein embedding tasks more efficiently. AVAILABILITY AND IMPLEMENTATION: The source code of MTDP is available via https://github.com/KennthShang/MTDP.

MOTIVATION: Bacteriophages (phages for short), which prey on and replicate within bacterial cells, have a significant role in modulating microbial communities and hold potential applications in treating antibiotic resistance. The advancement of high-throughput sequencing technology contributes to the discovery of phages tremendously. However, the taxonomic classification of assembled phage contigs still faces several challenges, including high genetic diversity, lack of a stable taxonomy system and limited knowledge of phage annotations. Despite extensive efforts, existing tools have not yet achieved an optimal balance between prediction rate and accuracy. RESULTS: In this work, we develop a learning-based model named PhaGenus, which conducts genus-level taxonomic classification for phage contigs. PhaGenus utilizes a powerful Transformer model to learn the association between protein clusters and support the classification of up to 508 genera. We tested PhaGenus on four datasets in different scenarios. The experimental results show that PhaGenus outperforms state-of-the-art methods in predicting low-similarity datasets, achieving an improvement of at least 13.7%. Additionally, PhaGenus is highly effective at identifying previously uncharacterized genera that are not represented in reference databases, with an improvement of 8.52%. The analysis of the infants' gut and GOV2.0 dataset demonstrates that PhaGenus can be used to classify more contigs with higher accuracy.

Accurately identifying potential drug-target interactions (DTIs) is a key step in drug discovery. Although many related experimental studies have been carried out for identifying DTIs in the past few decades, the biological experiment-based DTI identification is still timeconsuming and expensive. Therefore, it is of great significance to develop effective computational methods for identifying DTIs. In this paper, we develop a novel 'end-to-end' learning-based framework based on heterogeneous 'graph' convolutional networks for 'DTI' prediction called end-to-end graph (EEG)-DTI. Given a heterogeneous network containing multiple types of biological entities (i.e. drug, protein, disease, side-effect), EEG-DTI learns the low-dimensional feature representation of drugs and targets using a graph convolutional networks-based model and predicts DTIs based on the learned features. During the training process, EEG-DTI learns the feature representation of nodes in an end-to-end mode. The evaluation test shows that EEG-DTI performs better than existing state-of-art methods. The data and source code are available at: https://github.com/MedicineBiology-AI/EEG-DTI.

Deep learning techniques represented by deep convolutional neural networks (CNNs) have been widely used in polarimetric synthetic-aperture radar (PolSAR) image classification in recent years. One challenge is how to get a pleasant classification result with limited human-labeled samples. In this letter, a novel semisupervised classification method incorporating labels' semantic priors is proposed for PolSAR image classification with limited labeled samples. The core idea is that a good classification result map should have consistent regions and aligned boundaries. Thus, a cost function is proposed, which mainly contains three terms including a supervised term, a region consistency term, and a boundary kept term. The supervised term enforces the category label to be the same with the human-labeled labels. The region consistency term encourages the labels in one region to be consistent. The boundary-kept term constrains the region consistency term, preventing the classification map from being too smooth. An alternate iterative optimization method is proposed to solve this equation. First, a CNN is trained using the labeled samples and the classification probability map. Then, the classification probability map is updated by the prediction of the trained CNN and the labeled samples. Repeat these two procedures until the maximum number of iterations is met. Experiments on two real PolSAR images are conducted to validate the effectiveness of the proposed method compared with several state-of-the-art methods.