Ziheng Li, Liu Kang, Feng Xiao, Luxi Xing, Qingyi Si, Zhuoran Li, Weikang Gong, Deqing Yang, Yanghua Xiao, Hongcheng Guo. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Understanding RNA conformational dynamics is essential to understand its roles in complex biological processes. While computational methods have revolutionized the prediction of static 3D RNA structures, predicting local flexibility directly from structure remains a significant challenge. We developed DeepRMSF, a deep learning-based method that leverages atomic-level descriptions of RNA to predict vibrational flexibility given a tertiary structure. Trained on MD-derived root-mean-square fluctuations(RMSF), DeepRMSF was benchmarked on 371 nonredundant RNAs, with 311 RNAs used for five-fold cross-validation (PCC = 0.7219-0.7464) and 60 RNAs as an independent test set (PCC = 0.734), ensuring minimal sequence/structural similarity between sets. DeepRMSF predicts the local flexibility of medium-sized RNAs (~75 nucleotides) in ~8.2 s, achieving >3000-fold speed-up over MD simulations while maintaining strong extrapolative accuracy. Rather than replacing MD, DeepRMSF offers a scalable and practical alternative for transcriptome-scale screening of RNA flexibility, facilitating studies on RNA structure-dynamics-function relationships and supporting computational modeling in RNA biology.

MOTIVATION: Spatial transcriptomics (ST) profiles gene expression in tissue context, enabling spatial domain detection. However, relatively few methods jointly recover discrete spatial domains and continuous within-section pseudotemporal trends in a single framework. Current spatiotemporal approaches often emphasise trajectory continuity to recover smooth progression-associated gradients, but this may blur neighbouring domain boundaries and reduce clustering accuracy. Conversely, specialised spatial clustering algorithms typically rely on external single-cell trajectory tools rather than providing an integrated, spatially aware pseudotime model. RESULTS: We introduce SECTOR (Structural Entropy-based Clustering and pseudoTime ORdering), a lightweight deep graph learning framework that unifies spatial domain detection and pseudotime inference. SECTOR optimises a differentiable structural entropy (SE) objective on a fused spatial-expression graph, with spatial total variation regularisation to promote tissue continuity. Across seven benchmark datasets spanning standard and modern high-resolution ST platforms, SECTOR consistently outperformed existing spatiotemporal methods in clustering accuracy and matched or exceeded leading spatial clustering algorithms, while maintaining modest computational demands. In human breast cancer and mouse olfactory bulb case studies, SECTOR recovered spatially organised pseudotime patterns supported by semivariance, transition-gene, enrichment and marker-gene analyses. Together, these results show that SE-based learning provides an effective and scalable strategy for modelling within-section spatiotemporal organisation in ST. AVAILABILITY: SECTOR is available on GitHub at https://github.com/lhbcb/SECTOR and archived on Figshare at https://doi.org/10.6084/m9.figshare.32029830.

Large single-cell ribonucleic acid-sequencing (scRNA-seq) datasets offer unprecedented biological insights but present substantial computational challenges for visualization and analysis. While existing subsampling methods can enhance efficiency, they may not ensure optimal performance in downstream machine learning and deep learning (ML/DL) tasks. Here, we introduce scValue, a novel approach that ranks individual cells by 'data value' using out-of-bag estimates from a random forest model. scValue prioritizes high-value cells and allocates greater representation to cell types with higher variability in data value, effectively preserving key biological signals within subsamples. We benchmarked scValue on automatic cell-type annotation tasks across four large datasets, paired with distinct ML/DL models. Our method consistently outperformed existing subsampling methods, closely matching full-data performance across all annotation tasks. In three additional case studies-label transfer learning, cross-study label harmonization, and bulk RNA-seq deconvolution-scValue more effectively preserved T-cell annotations across human gut-colon datasets, more accurately reproduced T-cell subtype relationships in a human spleen dataset, and constructed a more reliable single-cell immune reference for cell-type deconvolution in simulated bulk tissue samples. Finally, using 16 public datasets ranging from tens of thousands to millions of cells, we evaluated subsampling quality based on computational time, Gini coefficient, and Hausdorff distance. scValue demonstrated fast execution, well-balanced cell-type representation, and distributional properties akin to uniform sampling. Overall, scValue provides a robust and scalable solution for subsampling large scRNA-seq data in ML/DL workflows. It is available as an open-source Python package installable via pip, with source code at https://github.com/LHBCB/scvalue.

Quality assessment for stitched panoramic images (SPIQA) is of great significance for the stitching algorithm optimization. By contrast, this task is much more challenging and arduous than traditional IQA task due to the high resolution of stitched panoramic images and the particularity and complexity of stitching distortions. For this task, we propose an effective method based on patch registration and bidimensional feature aggregation (PRBFA). First, inspired by the attention mechanism of the human visual system and the limited range of human vision, a soft patch segmentation and selection method is presented to determine the key patches in panoramic images to participate in the following patch matching and feature alignment stages, achieving patch registration between the panoramic image and the corresponding constituent images. Further, to fully simulate the human visual perception process from local viewport to panorama, the feature exploration is successively performed from local to global, which is also adaptive to the complexity of the distortions in stitched panoramic images. For performance testification, extensive experiments are conducted on the publicly released SPIQA database, the results of which prove the performance superiority of the PRBFA method.

Protein-deoxyribonucleic acid (DNA) interactions are important in a variety of biological processes. Accurately predicting protein-DNA binding affinity has been one of the most attractive and challenging issues in computational biology. However, the existing approaches still have much room for improvement. In this work, we propose an ensemble model for Protein-DNA Binding Affinity prediction (emPDBA), which combines six base models with one meta-model. The complexes are classified into four types based on the DNA structure (double-stranded or other forms) and the percentage of interface residues. For each type, emPDBA is trained with the sequence-based, structure-based and energy features from binding partners and complex structures. Through feature selection by the sequential forward selection method, it is found that there do exist considerable differences in the key factors contributing to intermolecular binding affinity. The complex classification is beneficial for the important feature extraction for binding affinity prediction. The performance comparison of our method with other peer ones on the independent testing dataset shows that emPDBA outperforms the state-of-the-art methods with the Pearson correlation coefficient of 0.53 and the mean absolute error of 1.11 kcal/mol. The comprehensive results demonstrate that our method has a good performance for protein-DNA binding affinity prediction. Availability and implementation: The source code is available at https://github.com/ChunhuaLiLab/emPDBA/.

Metal ion is an indispensable factor for the proper folding, structural stability and functioning of RNA molecules. However, it is very difficult for experimental methods to detect them in RNAs. With the increase of experimentally resolved RNA structures, it becomes possible to identify the metal ion-binding sites in RNA structures through in-silico methods. Here, we propose an approach called Metal3DRNA to identify the binding sites of the most common metal ions (Mg2+, Na+ and K+) in RNA structures by using a three-dimensional convolutional neural network model. The negative samples, screened out based on the analysis for binding surroundings of metal ions, are more like positive ones than the randomly selected ones, which are beneficial to a powerful predictor construction. The microenvironments of the spatial distributions of C, O, N and P atoms around a sample are extracted as features. Metal3DRNA shows a promising prediction power, generally surpassing the state-of-the-art methods FEATURE and MetalionRNA. Finally, utilizing the visualization method, we inspect the contributions of nucleotide atoms to the classification in several cases, which provides a visualization that helps to comprehend the model. The method will be helpful for RNA structure prediction and dynamics simulation study. Availability and implementation: The source code is available at https://github.com/ChunhuaLiLab/Metal3DRNA.

Data-driven discovery of image-derived phenotypes (IDPs) from large-scale multimodal brain imaging data has enormous potential for neuroscientific and clinical research by linking IDPs to subjects' demographic, behavioural, clinical and cognitive measures (i.e., non-imaging derived phenotypes or nIDPs). However, current approaches are primarily based on unsupervised approaches, without the use of information in nIDPs. In this paper, we proposed a semi-supervised, multimodal, and multi-task fusion approach, termed SuperBigFLICA, for IDP discovery, which simultaneously integrates information from multiple imaging modalities as well as multiple nIDPs. SuperBigFLICA is computationally efficient and largely avoids the need for parameter tuning. Using the UK Biobank brain imaging dataset with around 40,000 subjects and 47 modalities, along with more than 17,000 nIDPs, we showed that SuperBigFLICA enhances the prediction power of nIDPs, benchmarked against IDPs derived by conventional expert-knowledge and unsupervised-learning approaches (with average nIDP prediction accuracy improvements of up to 46%). It also enables the learning of generic imaging features that can predict new nIDPs. Further empirical analysis of the SuperBigFLICA algorithm demonstrates its robustness in different prediction tasks and the ability to derive biologically meaningful IDPs in predicting health outcomes and cognitive nIDPs, such as fluid intelligence and hypertension.

Panoramic image quality assessment (PIQA) is crucial to the successful application of technologies that can provide immersive visual experience. Stitching distortions are one of the main types of distortions that result in panoramic image degradation. However, most existing PIQA methods are general-purpose ones, which ignore the special characteristics of the stitching distortions caused by imperfect stitching algorithms. This results in unsatisfactory performance. To this end, we propose an effective stitched PIQA method, which consists of an imaginary reference generation (IRG) module and a hierarchical quality prediction (HQP) module. Among them, the IRG module is proposed to mimic the capability of the human visual system in imagining the raw version in the face of a degraded image. For the IRG module learning, we construct a large-scale database. The HQP module is presented to adapt to the particularity and complexity of stitching distortions, which is achieved by the pyramid feature aggregation. Extensive experiments and comparisons have been performed on the stitched PIQA database and the experimental results demonstrate the superiority of the proposed method in evaluating the quality of stitched panoramic images.

The three-dimensional (3D) chromosomal structure plays an essential role in all DNA-templated processes, including gene transcription, DNA replication and other cellular processes. Although developing chromosome conformation capture (3C) methods, such as Hi-C, which can generate chromosomal contact data characterized genome-wide chromosomal structural properties, understanding 3D genomic nature-based on Hi-C data remains lacking. Here, we propose a persistent spectral simplicial complex (PerSpectSC) model to describe Hi-C data for the first time. Specifically, a filtration process is introduced to generate a series of nested simplicial complexes at different scales. For each of these simplicial complexes, its spectral information can be calculated from the corresponding Hodge Laplacian matrix. PerSpectSC model describes the persistence and variation of the spectral information of the nested simplicial complexes during the filtration process. Different from all previous models, our PerSpectSC-based features provide a quantitative global-scale characterization of chromosome structures and topology. Our descriptors can successfully classify cell types and also cellular differentiation stages for all the 24 types of chromosomes simultaneously. In particular, persistent minimum best characterizes cell types and Dim (1) persistent multiplicity best characterizes cellular differentiation. These results demonstrate the great potential of our PerSpectSC-based models in polymeric data analysis.

MOTIVATION: The identification of binding hotspots in protein-RNA interactions is crucial for understanding their potential recognition mechanisms and drug design. The experimental methods have many limitations, since they are usually time-consuming and labor-intensive. Thus, developing an effective and efficient theoretical method is urgently needed. RESULTS: Here, we present SREPRHot, a method to predict hotspots, defined as the residues whose mutation to alanine generate a binding free energy change ≥2.0 kcal/mol, while others use a cutoff of 1.0 kcal/mol to obtain balanced datasets. To deal with the dataset imbalance, Synthetic Minority Over-sampling Technique (SMOTE) is utilized to generate minority samples to achieve a dataset balance. Additionally, besides conventional features, we use two types of new features, residue interface propensity previously developed by us, and topological features obtained using node-weighted networks, and propose an effective Random Grouping feature selection strategy combined with a two-step method to determine an optimal feature set. Finally, a stacking ensemble classifier is adopted to build our model. The results show SREPRHot achieves a good performance with SEN, MCC and AUC of 0.900, 0.557 and 0.829 on the independent testing dataset. The comparison study indicates SREPRHot shows a promising performance. AVAILABILITY AND IMPLEMENTATION: The source code is available at https://github.com/ChunhuaLiLab/SREPRHot. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Protein-RNA interactions play a critical role in various biological processes. The accurate prediction of RNA-binding residues in proteins has been one of the most challenging and intriguing problems in the field of computational biology. The existing methods still have a relatively low accuracy especially for the sequence-based ab-initio methods. RESULTS: In this work, we propose an approach aPRBind, a convolutional neural network-based ab-initio method for RNA-binding residue prediction. aPRBind is trained with sequence features and structural ones (particularly including residue dynamics information and residue-nucleotide propensity developed by us) that are extracted from the predicted structures by I-TASSER. The analysis of feature contributions indicates the sequence features are most important, followed by dynamics information, and the sequence and structural features are complementary in binding site prediction. The performance comparison of our method with other peer ones on benchmark dataset shows that aPRBind outperforms some state-of-the-art ab-initio methods. Additionally, aPRBind can give a better prediction for the modeled structures with TM-score≥0.5, and meanwhile since the structural features are not very sensitive to the refined 3D structures, aPRBind has only a marginal dependence on the accuracy of the structure model, which allows aPRBind to be applied to the RNA-binding site prediction for the modeled or unbound structures. AVAILABILITY AND IMPLEMENTATION: The source code is available at https://github.com/ChunhuaLiLab/aPRbind. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Non-coding RNA (ncRNA) and protein interactions play essential roles in various physiological and pathological processes. The experimental methods used for predicting ncRNA-protein interactions are time-consuming and labor-intensive. Therefore, there is an increasing demand for computational methods to accurately and efficiently predict ncRNA-protein interactions. RESULTS: In this work, we presented an ensemble deep learning-based method, EDLMFC, to predict ncRNA-protein interactions using the combination of multi-scale features, including primary sequence features, secondary structure sequence features, and tertiary structure features. Conjoint k-mer was used to extract protein/ncRNA sequence features, integrating tertiary structure features, then fed into an ensemble deep learning model, which combined convolutional neural network (CNN) to learn dominating biological information with bi-directional long short-term memory network (BLSTM) to capture long-range dependencies among the features identified by the CNN. Compared with other state-of-the-art methods under five-fold cross-validation, EDLMFC shows the best performance with accuracy of 93.8%, 89.7%, and 86.1% on RPI1807, NPInter v2.0, and RPI488 datasets, respectively. The results of the independent test demonstrated that EDLMFC can effectively predict potential ncRNA-protein interactions from different organisms. Furtherly, EDLMFC is also shown to predict hub ncRNAs and proteins presented in ncRNA-protein networks of Mus musculus successfully. CONCLUSIONS: In general, our proposed method EDLMFC improved the accuracy of ncRNA-protein interaction predictions and anticipated providing some helpful guidance on ncRNA functions research. The source code of EDLMFC and the datasets used in this work are available at https://github.com/JingjingWang-87/EDLMFC .

Neuroimaging allows for the non-invasive study of the brain in rich detail. Data-driven discovery of patterns of population variability in the brain has the potential to be extremely valuable for early disease diagnosis and understanding the brain. The resulting patterns can be used as imaging-derived phenotypes (IDPs), and may complement existing expert-curated IDPs. However, population datasets, comprising many different structural and functional imaging modalities from thousands of subjects, provide a computational challenge not previously addressed. Here, for the first time, a multimodal independent component analysis approach is presented that is scalable for data fusion of voxel-level neuroimaging data in the full UK Biobank (UKB) dataset, that will soon reach 100,000 imaged subjects. This new computational approach can estimate modes of population variability that enhance the ability to predict thousands of phenotypic and behavioural variables using data from UKB and the Human Connectome Project. A high-dimensional decomposition achieved improved predictive power compared with widely-used analysis strategies, single-modality decompositions and existing IDPs. In UKB data (14,503 subjects with 47 different data modalities), many interpretable associations with non-imaging phenotypes were identified, including multimodal spatial maps related to fluid intelligence, handedness and disease, in some cases where IDP-based approaches failed.

Deep learning has huge potential for accurate disease prediction with neuroimaging data, but the prediction performance is often limited by training-dataset size and computing memory requirements. To address this, we propose a deep convolutional neural network model, Simple Fully Convolutional Network (SFCN), for accurate prediction of brain age using T1-weighted structural MRI data. Compared with other popular deep network architectures, SFCN has fewer parameters, so is more compatible with small dataset size and 3D volume data. The network architecture was combined with several techniques for boosting performance, including data augmentation, pre-training, model regularization, model ensemble and prediction bias correction. We compared our overall SFCN approach with several widely-used machine learning models. It achieved state-of-the-art performance in UK Biobank data (N = 14,503), with mean absolute error (MAE) = 2.14y in brain age prediction and 99.5% in sex classification. SFCN also won (both parts of) the 2019 Predictive Analysis Challenge for brain age prediction, involving 79 competing teams (N = 2,638, MAE = 2.90y). We describe here the details of our approach, and its optimisation and validation. Our approach can easily be generalised to other tasks using different image modalities, and is released on GitHub.

BACKGROUND: The maintenance of protein structural stability requires the cooperativity among spatially neighboring residues. Previous studies have shown that conserved residues tend to occur clustered together within enzyme active sites and protein-protein/DNA interfaces. It is possible that conserved residues form one or more local clusters in protein tertiary structures as it can facilitate the formation of functional motifs. In this work, we systematically investigate the spatial distributions of conserved residues as well as hot spot ones within protein-RNA interfaces. RESULTS: The analysis of 191 polypeptide chains from 160 complexes shows the polypeptides interacting with tRNAs evolve relatively rapidly. A statistical analysis of residues in different regions shows that the interface residues are often more conserved, while the most conserved ones are those occurring at protein interiors which maintain the stability of folded polypeptide chains. Additionally, we found that 77.8% of the interfaces have the conserved residues clustered within the entire interface regions. Appling the clustering characteristics to the identification of the real interface, there are 31.1% of cases where the real interfaces are ranked in top 10% of 1000 randomly generated surface patches. In the conserved clusters, the preferred residues are the hydrophobic (Leu, Ile, Met), aromatic (Tyr, Phe, Trp) and interestingly only one positively charged Arg residues. For the hot spot residues, 51.5% of them are situated in the conserved residue clusters, and they are largely consistent with the preferred residue types in the conserved clusters. CONCLUSIONS: The protein-RNA interface residues are often more conserved than non-interface surface ones. The conserved interface residues occur more spatially clustered relative to the entire interface residues. The high consistence of hot spot residue types and the preferred residue types in the conserved clusters has important implications for the experimental alanine scanning mutagenesis study. This work deepens the understanding of the residual organization at protein-RNA interface and is of potential applications in the identification of binding site and hot spot residues.

MOTIVATION: Advances in neuroimaging and sequencing techniques provide an unprecedented opportunity to map the function of brain regions and identify the roots of psychiatric diseases. However, the results from most neuroimaging studies, i.e. activated clusters/regions or functional connectivities between brain regions, frequently cannot be conveniently and systematically interpreted, rendering the biological meaning unclear. RESULTS: We describe a brain annotation toolbox that generates functional and genetic annotations for neuroimaging results. The voxel-level functional description from the Neurosynth database and gene expression profile from the Allen Human Brain Atlas are used to generate functional/genetic information for region-level neuroimaging results. The validity of the approach is demonstrated by showing that the functional and genetic annotations for specific brain regions are consistent with each other; and further the region by region functional similarity network and genetic similarity network are highly correlated for major brain atlases. One application of brain annotation toolbox is to help provide functional/genetic annotations for newly discovered regions with unknown functions, e.g. the 97 new regions identified in the Human Connectome Project. Importantly, this toolbox can help understand differences between psychiatric patients and controls, and this is demonstrated using schizophrenia and autism data, for which the functional and genetic annotations for the neuroimaging changes in patients are consistent with each other and help interpret the results. AVAILABILITY AND IMPLEMENTATION: BAT is implemented as a free and open-source MATLAB toolbox and is publicly available at http://123.56.224.61:1313/post/bat. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.