Motivation Enhancer-promoter interactions (EPIs) are essential for gene regulation and disease progression. Recent studies have shown that distal enhancers can regulate target genes through interactions with nearby promoters, providing important insights into transcriptional regulation mechanisms. Although high-throughput experimental techniques have enabled large-scale identification of EPIs, these methods are often costly and time-consuming. In addition, existing computational approaches still face challenges in effectively integrating heterogeneous feature representations from different cell lines.Results We propose a stacked ensemble framework for EPI prediction that integrates feature representations from diverse cell line datasets using multiple machine learning algorithms. The extracted complementary patterns are further combined by an XGBoost classifier to improve robustness against overfitting. Experiments on six independent datasets show that the proposed method achieves superior accuracy and generalization compared with existing EPI prediction models, with an average AUROC of 0.909 while maintaining computational efficiency.Availability The source code and its archived release are available at GitHub and Zenodo. The Zenodo archive provides a versioned snapshot of the repository: https://zenodo.org/records/19952998

Single-cell 3D genome technologies provide unprecedented views of chromatin architecture, but the extreme sparsity and noise of contact maps limit robust detection of chromatin loops at the individual cell level. Here we present CellLoop, a computational framework for identifying chromatin loops from single-cell contact data by integrating intra-cellular and neighboring inter-cellular contacts through a density-based voting strategy. Applied to Dip-C data from the mouse brain, CellLoop achieves improved loop detection consistent with spatial distances and compartmentalization signals, revealing single cell-specific chromatin loops associated with transcriptional regulation and cell identity. In HiRES embryogenesis data, CellLoop enables finer cell subtype delineation by reducing confounding cell cycle effects. Integration with GAGE-seq and MERFISH data redefines spatial domain functions through chromatin loop dynamics. Together, CellLoop provides a scalable and accurate approach for characterizing chromatin loop variability at single-cell resolution and highlights the utility of 3D genome features in interpreting transcriptional and spatial heterogeneity.

Neural stem cells (NSCs) play a crucial role in neural regeneration following spinal cord injury (SCI) owing to their self-proliferative and multidirectional differentiation capabilities. This study examined the biological properties of adult spinal cord-derived NSCs (sp-NSCs) and the role of Notch receptors in regulating their activation. NSCs were isolated from the spinal cords of 8-week-old C57/BL6 mice, and their biological properties, including the gene expression profile of Notch receptors, were subsequently analyzed using single-cell RNA sequencing (scRNA-Seq) and bioinformatics. The NSCs were subsequently infected with lentiviral vectors encoding Notch1 shRNA, Notch2 shRNA, and a combination of Notch1 and Notch2 shRNA sequences, and evaluated using Sphere assays and EdU staining to determine their activation effects. The expression levels of downstream genes, NICD and Rbpj, in the Notch signaling pathway, as well as the related target genes, Hes1 and Hey1, were subsequently quantified using Western blot analysis. Intact adult mouse sp-NSCs predominantly existed in a quiescent state, with their population increasing significantly with age. Notch receptors served as critical regulators of adult sp-NSC activation. Notably, Notch1 expression was significantly elevated compared to Notch2 and Notch3, demonstrating its predominant role in sustaining NSC activation. In contrast, Notch2 and Notch3 were primarily responsible for maintaining NSCs in a quiescent state. Overall, both Notch1 and Notch2 signals are involved in different regulatory roles that facilitate the activation and fate determination of NSCs via NICD-Rbpj.

Motivation The rapid development of spatial multi-omics technology enables the simultaneous measurement of gene and protein expression alongside spatial location, providing valuable insights into tissue heterogeneity. However, challenges such as low spatial resolution and high feature dimensionality complicate data integration and biological interpretation.Results To address these issues, we propose SpaMFG, an innovative feature-group-level framework for interpretable spatial multi-omics integration. SpaMFG leverages spatial location information and introduces a spatial proximity weighting method to improve feature grouping accuracy. Additionally, it employs a new cross-omics feature group matching method that combines spatial location and Jaccard similarity to construct a weighted cost matrix, which is optimized using the Hungarian algorithm. This approach enhances the biological interpretability of cross-omics feature relationships. We evaluated SpaMFG's performance through comparative analysis on the human lymph node dataset, demonstrating its effectiveness. Further applications on human tonsils, mouse spleens, and mouse thymus datasets confirmed the robustness of SpaMFG in various biological contexts.Availability and implementation The source code for SpaMFG is available at https://github.com/LiangYu-Xidian/SpaMFG.

Spatial transcriptomics (ST) profiles gene expression with spatial context, but most platforms capture multicellular spots containing mixed cell types, making accurate deconvolution essential. Existing reference-based methods using scRNA-seq often ignore spatial dependency and gene-level contribution, yielding fragmented maps and limited insight into domain-specific programs. Here, we propose a gene-aware heterogeneous graph attention network called STGnet for spatial transcriptomics deconvolution and functional annotation. Leveraging a hybrid pseudo-spot generation strategy that captures realistic spatially enriched cell-type patterns, STGnet accurately integrates spatial adjacency, transcriptional similarity, and gene-spot associations within a unified heterogeneous network. Attention weights highlight domain-specific genes for interpretable domain annotation. Importantly, STGnet can characterize spatially ordered functional programs across domains that may be associated with disease progression. These insights may facilitate the discovery of spatial disease mechanisms and improve understanding of pathological tissue organization. Experiments on simulated and real datasets show that STGnet achieves the best overall performance compared with state-of-the-art methods.

Abstract Multiple sequence alignment (MSA) is a fundamental problem in computational bioinformatics, playing a critical role in genome biology, especially in long read sequencing and assembly. One solution for representing and solving MSA is Partial Order Alignment (POA), which employs Directed Acyclic Graphs (DAGs) to represent sequence relationships. However, when facing the ultra-long, error-prone reads (e.g., >100 kbps), existing POA algorithms with quadratic space complexity become impractical due to excessive memory consumption. This paper introduces the linearPOA, which based on divide-and-conquer strategy to solve the POA, aimed at saving memory compared to quadratic space complexity algorithms like SPOA, abPOA and TSTA. Particularly notable is its capability to save up to 102.74 times memory usage when aligning sequences with 100 kbp reads, compared to the abPOA method using non-heuristic methods. The algorithm was implemented within the linearPOA library, providing functionality for POA and foundational support for sequencing analysis, like error correction for reads. The linearPOA algorithm provides memory-efficient algorithms for long-read sequencing, especially in directly assembling long reads like 100 kbp reads. Availability The linearPOA library is freely available at https://github.com/malabz/linearPOA , and the data underlying this article are available in Zenodo, at https://doi.org/10.5281/zenodo.15637837 . Supplementary information Supplementary information are available at BioRxiv online.

The rapid expansion of whole-genome sequencing (WGS) has highlighted the important contribution of noncoding variants to human disease, yet their pathogenic mechanisms remain difficult to resolve. Traditional statistical and experimental approaches often struggle to capture complex regulatory interactions or establish causal links, leaving many noncoding variants classified as variants of uncertain significance in clinical databases. Recent advances in computational modeling have substantially improved pathogenicity prediction by integrating genomic, epigenetic, and structural information. In parallel, genome language model (gLM)-inspired methods have enabled more context-aware interpretation of noncoding sequences and improved model generalization. This review summarizes current computational approaches, data modalities, and evaluation strategies for noncoding variant pathogenicity prediction, discusses key challenges in interpretability and data heterogeneity, and highlights emerging opportunities for clinical translation.

BACKGROUND:Multiple sequence alignment (MSA) continues to be a central challenge in comparative genomics, where the quality of alignment plays a crucial role in determining the accuracy of downstream analyses. However, the challenge of large-scale alignment remains significant. FINDINGS:This article introduces deMEM, a novel and effective framework for DNA multiple sequence alignment, which enables existing MSA methods such as MAFFT to handle extremely large sequences. deMEM is a 3-stage alignment process: (i) representing maximum exact matches using a de Bruijn graph and clustering them based on their area, (ii) employing a novel divide-and-conquer framework for alignment, and (iii) providing profile-profile alignment between different clusters. CONCLUSIONS:DeMEM enables existing methods like MAFFT to align an extremely large number of sequences, including long sequences that cannot be directly aligned, such as those in a dataset of a thousand monkeypox virus genomes. The deMEM package is free and available at https://github.com/malabz/deMEM.

BackgroundN6-methyladenosine (m6A), the most predominant post-transcriptional RNA modification, regulates splicing, translation, and decay processes. Its dysregulation is implicated in cancers, metabolic disorders, and neurological diseases. Despite accumulating evidence highlighting m6A as a key player in human pathologies, no previous computational framework has investigated the high-order associations among m6A sites, diseases, and drugs within a unified model.ResultsHere, we introduce HNRM, a data-driven approach designed to model hyperedges across these entities. We frame this problem as a high-order link-prediction task on a hypergraph. We employ a hypergraph neural network based on hyperedge neighborhoods to learn embedding representations of both hyperedges and nodes.ConclusionsThe performance of HNRM is evaluated on a newly collected and processed m6A dataset, as well as on five additional datasets from other domains, demonstrating its superior effectiveness. Ablation studies and Gene Ontology enrichment analysis further validate its capability in identifying potential associations.

Motivation Predicting drug-induced transcriptional perturbations is critical for precision medicine, yet existing models fail to capture multimodal biological context, limiting generalization across unseen drugs and cell lines.Results We present PertDiff, a conditional diffusion framework that integrates control gene expression, LLM-derived cell semantics, and pretrained molecular graph representations to predict transcriptome-wide perturbations. PertDiff outperforms state-of-the-art baselines in prediction accuracy and generalizes robustly across drugs and cell lines. It further demonstrates translational utility through accurate drug sensitivity prediction, therapeutic repurposing for pancreatic cancer, and concordance with real-world clinical treatment outcomes, establishing it as a biologically grounded transcriptomic modeling tool.Availability The source code and data are available at https://github.com/Panda-myj/PertDiff and https://doi.org/10.5281/zenodo.18427848.

BACKGROUND:RNA post-transcriptional modifications involve the addition of chemical groups to RNA molecules or alterations to their local structure. These modifications can change RNA base pairing, affect thermal stability, and influence RNA folding, thereby impacting alternative splicing, translation, cellular localization, stability, and interactions with proteins and other molecules. Accurate prediction of RNA modification sites is essential for understanding modification mechanisms. RESULTS:We propose a novel deep learning model, YModPred, which accurately predicts multiple types of RNA modification sites in S. cerevisiae based on RNA sequences. YModPred combines convolution and self-attention mechanisms to enhance the model's ability to capture global sequence information and improve local feature learning. The model can predict multi-type RNA modification sites. Comparative analysis against benchmark models demonstrates that YModPred outperforms existing state-of-the-art methods in predicting various RNA modification types. Additionally, the model's prediction performance is further validated through visualization and motif analysis. CONCLUSIONS:YModPred is a deep learning-based model that effectively captures sequence features and dependencies, enabling accurate prediction of multi-type RNA modification sites in S. cerevisiae. We believe it will facilitate further research into the mechanisms of RNA modifications.

These open problems were presented in the Problem Sessions held during the Tianyuan Workshop on Computability Theory and Descriptive Set Theory, June 16-20, 2025. The problems are organized into sections named after their contributors, in the order of their presentations during the workshop. Notes were taken and compiled by Wei Dai, Feng Li, Ruiwen Li, Ming Xiao, Xu Wang, Víctor Hugo Yañez Salazar, and Yang Zheng.

Trajectory inference algorithms based on single-cell omics data are powerful tools for predicting and dissecting cell differentiation. However, most existing tools are tailored to specific datasets and lack generalizability across diverse data types. To ad-dress this limitation, we developed CellFateExplorer, systematically evaluates the performance of x trajectory inference methods across y datasets. Through an interactive web interface, CellFateExplorer provides guidance on method selection and down-stream analysis for specific datasets. In summary, CellFateExplorer is an integrated platform for exploring cell fate. ### Competing Interest Statement The authors have declared no competing interest.

Drug-target interaction (DTI) prediction is crucial for drug discovery, as it accelerates candidate screening and reduces development costs. However, existing computational methods are often limited to a single perspective and cannot simultaneously consider the biological information and complex associations of drugs and targets. Although multimodal data have been introduced, the complementarity and interaction of multi-source information remain underutilized, making efficient multi-view feature fusion a key challenge. In this paper, we propose a DTI prediction framework based on multi-view feature fusion and contrastive learning, named MFCL-DTI. It integrates sequence feature as well as structural and semantic information of heterogeneous graph. A multi-view adaptive fusion module facilitates cross-view feature fusion, while multi-view contrastive learning enhances feature representation. Experimental results demonstrate that MFCL-DTI outperforms existing methods, validating its effectiveness in DTI prediction.

In recent years, many approved drugs have been discovered using phenotypic screening, which elaborates the exact mechanisms of action or molecular targets of drugs. Drug susceptibility prediction is an important type of phenotypic screening. Large-scale pharmacogenomics studies have provided us with large amounts of drug sensitivity data. By analyzing these data using computational methods, we can effectively build models to predict drug susceptibility. However, due to the differences in data distribution among databases, researchers cannot directly utilize data from multiple sources. In this study, we propose a deep transfer learning model. We integrate the genomic characterization of cancer cell lines with chemical information on compounds, combined with the Encyclopedia of Cancer Cell Lines (CCLE) and the Genomics of Cancer Drug Sensitivity (GDSC) datasets, through a domain-adapted approach and predict the half-maximal inhibitory concentrations (IC50 values). Afterward, the validity of the prediction results of our model is verified. This study effectively addresses the challenge of cross-database distribution discrepancies in drug sensitivity prediction by integrating multi-source heterogeneous data and constructing a deep transfer learning model. This model serves as a reliable computational tool for precision drug development. Its widespread application can facilitate the optimization of therapeutic strategies in personalized medicine while also providing technical support for high-throughput drug screening and the discovery of new drug targets.

As a remote regulatory element of DNA, enhancers play a pivotal role in embryonic development, regulating gene expression, homeostasis and disease occurrence in a variety of biological processes. Identifying enhancers facilitates the exploration of biological processes and mechanisms. However, experimental methods utilized to identify enhancers are high time investment and expensive. Currently, many computational methods have been developed for large-scale enhancer identification. In this research, a new computational method called MC-MKHFIS was proposed to identify enhancers. The method is built upon kernelized high-order fuzzy inference system (KHFIS) and incorporates multiple kernel learning (MKL) techniques and mixture correntropy (MC) loss. We use MKL to identify the feature space that is conducive to sample representation, which is then used to construct the fuzzy kernel matrix. To further strengthen the performance and robustness of the method, we employ MC to reconstruct the objective function of the model. From the experimental results reported on two enhancer datasets our method demonstrates excellent classification performance.

MOTIVATION:Enhancers and promoters, as regulatory DNA elements, play pivotal roles in gene expression, homeostasis, and disease development across various biological processes. With advancing research, it has been uncovered that distal enhancers may engage with nearby promoters to modulate the expression of target genes. This discovery holds significant implications for deepening our comprehension of various biological mechanisms. In recent years, numerous high-throughput wet-lab techniques have been created to detect possible interactions between enhancers and promoters. However, these experimental methods are often time-intensive and costly. RESULTS:To tackle this issue, we have created an innovative deep learning approach, EPIPDLF, which utilizes advanced deep learning techniques to predict EPIs based solely on genomic sequences in an interpretable manner. Comparative evaluations across six benchmark datasets demonstrate that EPIPDLF consistently exhibits superior performance in EPI prediction. Additionally, by incorporating interpretable analysis mechanisms, our model enables the elucidation of learned features, aiding in the identification and biological analysis of important sequences. AVAILABILITY AND IMPLEMENTATION:The source code and data are available at: https://github.com/xzc196/EPIPDLF.

Three-dimensional molecular generation is critical in drug design. However, current methods often rely on point clouds or oversimplified interaction models, limiting their ability to accurately represent molecular structures. To address these challenges, this paper proposes the multiscale graph equivariant diffusion model for 3D molecule design (MD3MD). MD3MD partitions molecular conformations into multiscale graphs, assigning different weights to capture atomic interactions across scales. This framework guides the diffusion process, enabling high-quality 3D molecular generation. Experimental results demonstrate that MD3MD excels in both unconditional and conditional generation tasks, producing diverse, stable, and innovative molecules that meet specified conditions. Visualization highlights MD3MD’s ability to learn domain-specific patterns and generate molecules distinct from existing datasets while maintaining distributional consistency. By effectively exploring chemical space, MD3MD surpasses previous methods in generating innovative and chemically diverse molecules, offering a notable advancement in the field of molecular design.

G-protein-coupled receptors (GPCRs) are important protein molecules in the field of cell signaling and are widely found in various organisms. GPCRs play an important role in a variety of physiological processes and are important drug targets for a variety of diseases. Accurate prediction of GPCRs using machine learning is useful for drug design in a variety of related diseases. In this paper, we propose a method for identifying GPCRs based on mixed-feature vectors. We combine three individual features, such as 400D, N-gram and Parallel correlation pseudo amino acid composition (PC-PseAAC), using mixed-feature representation methods, which are evaluated by Random Forest, Naove Bayes, and J48 for classification purposes. To measure the performance of this classifier, ten-fold cross-validation is used. Two dimensionality reduction methods—the max-relevance-max-distance (MRMD) and t-Distributed Stochastic Neighbor Embedding (t-SNE)—are applied to reduce the feature dimension. The 400D and PC-PseAAC feature extraction methods are combined, the random forest is used as the classifier, and the area under the curve (AUC) is up to 0.9413. Therefore, among these methods, the new feature vector obtained by combining the two features shows the best performance, and the mixed feature is better than the single feature.

Spatial transcriptomics (STs) has emerged as a transformative approach to elucidate cellular heterogeneity and spatial organization within complex tissue microenvironments. However, the analysis of ST data is challenged by limited spatial resolution, resulting in mixed expression profiles at each spatial location. Moreover, the precious spatial information is rarely exploited, and noise issues in spatial transcriptomes (STs) are often overlooked by computational deconvolution methods. In this study, a novel computational framework for STs deconvolution (DeCoST), called DeCoST, is presented. DeCoST capitalizes on the valuable spatial context information by integrating a Gaussian kernel-based conditional autoregressive model. Additionally, the method employs domain adaptation techniques to address platform effects between single-cell and ST data, enabling robust cell type identification. Evaluations on simulated datasets under diverse spatial configurations, as well as real-world case studies on human pancreatic ductal adenocarcinoma, mouse olfactory bulb, and mouse brain samples, demonstrate the superior performance of DeCoST compared to existing deconvolution approaches. The method's ability to accurately map region-specific cell types and uncover spatial interactions advances our understanding of complex tissue organization and function, with broad applications in disease research and developmental biology.