Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Graph Neural Networks”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources
PubMed2005

Three-stage prediction of protein beta-sheets by neural networks, alignments and graph algorithms.

MOTIVATION: Protein beta-sheets play a fundamental role in protein structure, function, evolution and bioengineering. Accurate prediction and assembly of protein beta-sheets, however, remains challenging because protein beta-sheets require formation of hydrogen bonds between linearly distant residues. Previous approaches for predicting beta-sheet topological features, such as beta-strand alignments, in general have not exploited the global covariation and constraints characteristic of beta-sheet architectures. RESULTS: We propose a modular approach to the problem of predicting/assembling protein beta-sheets in a chain by integrating both local and global constraints in three steps. The first step uses recursive neural networks to predict pairing probabilities for all pairs of interstrand beta-residues from profile, secondary structure and solvent accessibility information. The second step applies dynamic programming techniques to these probabilities to derive binding pseudoenergies and optimal alignments between all pairs of beta-strands. Finally, the third step uses graph matching algorithms to predict the beta-sheet architecture of the protein by optimizing the global pseudoenergy while enforcing strong global beta-strand pairing constraints. The approach is evaluated using cross-validation methods on a large non-homologous dataset and yields significant improvements over previous methods. AVAILABILITY: http://www.igb.uci.edu/servers/psss.html.

Algorithms↗
PubMed1996

Neural network for dynamic binding with graph representation: form, linking, and depth-from-occlusion.

A neural network is presented that explicitly represents form attributes and relations between them, thus solving the binding problem without temporal coding. Rather, the network create a graph representation by dynamically allocating nodes to code local form attributes and establishing arcs to link them. With this representation, the network selectively groups and segments in depth objects based on line junction information, producing results consistent with those of several recent visual search experiments. In addition to depth-from-occlusion, the network provides a sufficient framework for local line-labeling processes to recover other three-dimensional (3-D) variables, such as edge/surface contiguity, edge slant, and edge convexity.

Form Perception↗
PubMed2026

GenOT: generative optimal transport enables spatiotemporal interpolation and generation in cross-platform spatial transcriptomics.

Spatial transcriptomics technologies have revolutionized the analysis of spatial gene expression, yet integrating spatial information and generating data across heterogeneous samples remain challenging. We present GenOT, a generative framework combining multi-scale graph self-supervised contrastive learning with optimal transport barycenter theory for efficient cross-slice and cross-platform spatiotemporal interpolation. The core innovation of GenOT lies in introducing an optimal transport barycenter-based interpolation algorithm, which mathematically models spatial distribution differences across heterogeneous samples to reconstruct spatiotemporal gene expression dynamics. Extensive evaluations demonstrate that GenOT consistently outperforms existing approaches in spatial domain identification, cross-platform interpolation, and developmental trajectory reconstruction.

Spatial Transcriptomics↗
PubMed2026

CAGNet: a structure-aware clustering-alternated graph network for cell-cell interaction inference in spatial transcriptomics.

MOTIVATION: Understanding cell-cell interactions (CCIs) in spatial transcriptomics is crucial for uncovering the spatial organization and functional heterogeneity of tissues. However, existing graph-based models typically rely on static clustering or fixed adjacency structures, which limits their ability to capture dynamic cellular relationships. RESULTS: We propose CAGNet, a two-stage framework for CCI inference from spatial transcriptomics data. In Stage 1, a Graph Attention Network encoder with joint feature and graph reconstruction learns structure-aware node embeddings from spatial gene expression profiles. In Stage 2, an alternating optimization mechanism iteratively updates cluster centers via KL-guided soft assignment and refines node embeddings through spatial graph reconstruction, establishing a closed-loop between representation learning and clustering. Experiments on three 10x Genomics Visium datasets demonstrate that CAGNet consistently outperforms six CCI inference baselines across ACC, AUC, AP, Precision, Recall, and F1. CAGNet also achieves the highest Adjusted Rand Index on all three datasets against six spatial domain identification methods, confirming that the learned embeddings capture biologically relevant spatial organization. Information-theoretic analysis further shows that CAGNet retains the highest mutual information between input features and learned embeddings among all compared methods. Ablation studies and 5-fold cross-validation confirm the contribution of each component and the reproducibility of the results. AVAILABILITY: The proposed method is implemented in the CAGNet package available at http://github.com/mahan1233333-maker/CAGNet .

Spatial Transcriptomics↗
PubMed2006

Recursive processing of cyclic graphs.

Recursive neural networks are a powerful tool for processing structured data. According to the recursive learning paradigm, the input information consists of directed positional acyclic graphs (DPAGs). In fact, recursive networks are fed following the partial order defined by the links of the graph. Unfortunately, the hypothesis of processing DPAGs is sometimes too restrictive, being the nature of some real-world problems intrinsically cyclic. In this paper, a methodology is proposed, which allows us to process any cyclic directed graph. Therefore, the computational power of recursive networks is definitely established, also clarifying the underlying limitations of the model.

Journal Article↗
PubMed2026

GiGCN: a network-based framework for uncovering synthetic lethal and viable genetic interactions.

Genetic interactions (GIs) underpin the functional connectivity of genes and pathways, and are important for dissecting genotype-phenotype relationships and identifying therapeutic targets for diseases. However, the scale of the human genome restricts systematic experimental interrogation of GIs. Existing computational tools focus on predicting synthetic lethality (SL) and synthetic viability (SV), the two primary forms of GIs, yet their accuracy and biological interpretability are compromised by inadequate modeling of the molecular mechanisms behind positive and negative interactions, as well as the limitation of negative samples. To overcome these challenges, we developed Genetic Interaction Graph Convolutional Network (GiGCN), a signed network modeling framework for the joint identification of gene pairs with SL and SV. We built a high-confidence signed genetic network by integrating verified GIs, and non-interacting gene pairs, together with gene semantic similarity derived from biological processes. By leveraging disentangled subspace decomposition, this framework separately models distinct functional dimensions within gene networks, enabling robust representation of context-dependent regulatory relationships and accurate discrimination of SL and SV events. Benchmark experiments demonstrate that GiGCN outperforms state-of-the-art approaches (area under receiver operating-characteristic curve: 0.978, and area under precision-recall curve: 0.944). Further analyses reveal biologically meaningful insights, including known and novel SL interactions centered on the oncogene MYC Proto-Oncogene (MYC), as well as SV interactions linked to autophagy and mitophagy pathways. This study provides a robust and interpretable network-based strategy for systematically exploring GIs. The GiGCN framework not only improves the precision of SL and SV prediction, but also offers mechanistic insights into gene functional relationships, thereby supporting the discovery of actionable therapeutic targets for cancer and other human diseases.

Humans↗
PubMed2023

Towards mechanistic models of mutational effects: Deep learning on Alzheimer's Aβ peptide.

Deep Mutational Scanning (DMS) has enabled multiplexed measurement of mutational effects on protein properties, including kinematics and self-organization, with unprecedented resolution. However, potential bottlenecks of DMS characterization include experimental design, data quality, and depth of mutational coverage. Here, we apply deep learning to comprehensively model the mutational effect of the Alzheimer's Disease associated peptide Aβ42 on aggregation-related biochemical traits from DMS measurements. Among tested neural network architectures, Convolutional Neural Networks and Recurrent Neural Networks are found to be the most cost-effective models with high performance even under insufficiently-sampled DMS studies. While sequence features are essential for satisfactory prediction from neural networks, geometric-structural features further enhance the prediction performance. Notably, we demonstrate how mechanistic insights into phenotype may be extracted from the neural networks themselves suitably designed. This methodological benefit is particularly relevant for biochemical systems displaying a strong coupling between structure and phenotype such as the conformation of Aβ42 aggregate and nucleation, as shown here using a Graph Convolutional Neural Network (GCN) developed from the protein atomic structure input. In addition to accurate imputation of missing values (which here ranged up to 55% of all phenotype values at key residues), the mutationally-defined nucleation phenotype generated from a GCN shows improved resolution for identifying known disease-causing mutations relative to the original DMS phenotype. Our study suggests that neural network derived sequence-phenotype mapping can be exploited not only to provide direct support for protein engineering or genome editing but also to facilitate therapeutic design with the gained perspectives from biological modeling.

Alzheimer's disease↗
PubMed2026

ORBIT: Oncogenic Representation Learning via Bi-Prototype Contrastive Learning in Hyperbolic Space for cancer driver gene identification.

Accurate identification of cancer driver genes is crucial for precision oncology but remains challenging due to the complexity of integrating heterogeneous data and modeling dynamic biological systems. To address these limitations, we propose ORBIT (Oncogenic Representation Learning via Bi-Prototype Contrastive Learning in Hyperbolic Space). Our framework synergistically fuses multi-omics profiles with functional network data using a context-adaptive graph reweighting mechanism to capture cancer-specific dynamics. The model employs a bi-prototype contrastive learning strategy within hyperbolic space, which aligns gene representations around distinct driver and non-driver semantic anchors while preserving the intrinsic hierarchy of biological networks. Comprehensive evaluations demonstrate that ORBIT achieves highly competitive stability in pan-cancer analysis while consistently outperforming state-of-the-art methods in cancer-specific predictions. Furthermore, functional enrichment analysis confirms that the model effectively segregates core cancer pathways, and drug sensitivity profiling validates the clinical relevance of the identified drivers. By integrating hyperbolic geometry with context-adaptive learning, ORBIT offers a robust and interpretable paradigm for precision medicine. The source codes and datasets are publicly accessible at https://github.com/spcho-dev/ORBIT.

Humans↗
PubMed2026

PEARL: integrative multi-omics classification and omics feature discovery via deep graph learning.

MOTIVATION: Integrating multi-omics data provides valuable insights into biological processes by capturing information across multiple molecular layers, enabling a comprehensive understanding of complex diseases and driving advancements in precision medicine. However, existing computational methods for multi-omics integration face significant challenges, such as low reliability and poor generalizability, due to the high dimensionality and low sample size nature of omics data. RESULTS: To address these challenges, we present PEARL (Pearson-Enhanced spectrAl gRaph convoLutional networks), a novel deep graph learning method for biomedical classification and functional important omics features identification. PEARL leverages a simple yet effective learning architecture to achieve superior and robust performance in high-dimensional, low-sample-size multi-omics settings. Our results demonstrate that PEARL significantly outperforms existing state-of-the-art methods on both synthetic and real biomedical datasets. Furthermore, applied to Alzheimer's disease (AD) brain multi-omics data, features prioritized by PEARL lead to functionally important genes that demonstrate significant enrichment in AD-related pathways. These findings highlight PEARL's practical utility in biomedical research and its potential to enhance biological interpretability in multi-omics studies. AVAILABILITY AND IMPLEMENTATION: The source code of our computational framework is available at https://github.com/zqq121017/PEARL.

Multiomics↗
PubMed2026

GraphyloVar: predicting the impact of non-coding variants using a multi-species sequence model.

MOTIVATION: Understanding the functional impact of genetic variants is a key problem for precision medicine. Tools like CADD, PhyloP, and PhastCons are useful, but they often look at each position in the genome in isolation. This means they can miss important information from the evolutionary history that connects different species. In this paper, we extend our previous model, Graphylo, to predict the effects of variants. Our new model, GraphyloVar, is built to directly utilize the phylogenetic tree that relates the species. RESULTS: GraphyloVar is a deep learning model that considers both DNA sequence and evolutionary patterns from many species. It uses two main components: Graph Convolutional Networks (GCNs) to process the phylogenetic tree, and Transformer encoders to extract features from the DNA sequences. Pre-trained to predict population-level allele frequencies on the TOPMed whole-genome sequencing cohort, GraphyloVar achieves an AUROC of 0.6246 zero-shot on &#x223c;149M held-out variants, and an ensemble with CADD reaches 0.6442 (+0.020, P<10-15). Fine-tuned GraphyloVar achieves the highest AUROC across all 13 MPRA benchmark datasets. By integrating deep learning with explicit phylogenetic input, GraphyloVar offers a powerful and complementary approach to variant effect prediction that utilizes the full evolutionary history from many species to better identify and prioritize important non-coding variants. AVAILABILITY AND IMPLEMENTATION: Code and datasets are available at https://github.com/DongjoonLim/GraphyloVar under DOI: 10.5281/zenodo.20616818.

Phylogeny↗
PubMed2026

A nonlinear multi-omics data integration and classification model based on pathway self-attention and graph convolutional networks.

The abundance of omics data has significantly advanced the development of multi-omics data integration techniques. Non-linear embedding approaches for data integration have gradually become the mainstream in multi-omics research, as these approaches can substantially improve cancer analysis by enhancing the quality of the embeddings. However, current multi-omics data integration methods are typically confined to omics measurements, neglecting domain-specific prior knowledge encompassing biological pathways. In this study, we proposed a multi-omics integrated classification model, PathTransGCN, based on pathway self-attention and graph convolutional networks (GCN). The model integrated biological pathway information into multi-omics data analysis with the aim of enhancing the accuracy of cancer classification. Multi-omics data for breast cancer (BRCA), non-small cell lung cancer (NSCLC), and low-grade glioma (LGG) were obtained from The Cancer Genome Atlas (TCGA) and UCSC Xena databases. These data included gene mutations, DNA methylation, copy number variations, and gene expression, and were used to assess the model's generalizability across different cancers. First, PathTransGCN employed a pathway self-attention module to learn latent representations of samples across different pathways, thereby obtaining multi-omics integration vectors. Concurrently, a patient similarity network (PSN) was constructed using the similarity network fusion (SNF) approach. Second, the integrated vectors and the PSN were jointly fed into a GCN for end-to-end training, enabling precise classification of cancer subtypes. Through multi-omics data analysis of the BRCA dataset, PathTransGCN outperformed several popular algorithms (such as MoGCN and DeePathNet) in the five-class classification of cancer subtypes, achieving an accuracy rate of 87.6% and an F1 score of 86.4%. Moreover, the model demonstrated robust generalization capabilities across both NSCLC and LGG datasets, while effectively identifying key disease-associated biomarkers at the pathway level. Experimental results demonstrate that PathTransGCN exhibits outstanding performance in integrating omics data and delivering interpretable classification outcomes, presenting significant potential for clinical applications.

Humans↗
PubMed2026

Artificial Intelligence-Driven Multi-Omics Analysis Reveals Hydroxytyrosol Targeting of the TXNIP-NLRP3 Inflammasome Axis in Traumatic Brain Injury.

Traumatic brain injury (TBI) induces secondary neuroinflammation driven by oxidative stress, inflammasome activation, and immune remodeling, yet specific mechanism-guided pharmacological interventions remain limited. This study established an artificial intelligence (AI)-integrated network pharmacology and multi-omics framework to evaluate whether hydroxytyrosol (HT), an olive-derived natural polyphenol, may regulate TBI-related neuroinflammatory targets centered on the TXNIP/NLRP3 inflammasome axis. Starting from the SMILES structure of HT, potential targets were predicted using PharmMapper, SwissTargetPrediction, and the Similarity Ensemble Approach and were standardized to UniProt identifiers. TBI-associated genes were integrated from GeneCards, DisGeNET, OMIM, and the Therapeutic Target Database. The overlapping target set was analyzed using STRING-based protein-protein interaction (PPI) networks, MCODE, CytoHubba, Gene Ontology (GO), and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment. Public GEO transcriptomic datasets (GSE123831 and GSE104687) were used for cross-platform expression validation, differential expression analysis, and exploratory CIBERSORT-based immune infiltration estimation. Random forest (RF), multilayer perceptron (MLP), graph convolutional network (GCN), graph attention network (GAT), SHAP/LIME explainability analysis, LASSO inflammatory-risk scoring, and two-sample Mendelian randomization (MR) were further applied for target prioritization, immune phenotype mapping, and genetic association analysis. Seventy-three overlapping HT-TBI targets were identified. PPI and topology analyses prioritized TXNIP, NLRP3, CASP1, MAPK1, and TP53 as key hubs enriched in inflammasome activation, oxidative stress, apoptosis, and NOD-like receptor signaling. TXNIP, NLRP3, and CASP1 were consistently upregulated in both TBI transcriptomic datasets. LM22-based immune deconvolution suggested increased pro-inflammatory immune signatures and a positive TXNIP-M1 macrophage association (r&#x202f;=&#x202f;0.63, p < 0.001), which should be interpreted as a transcriptome-derived hypothesis rather than validated murine immune-cell proportions. AI-based models consistently ranked TXNIP/NLRP3 as high-contribution features under internal validation, and removal of these targets reduced model performance. A five-gene inflammatory score achieved an internally evaluated AUC of 0.87, while two-sample MR supported positive genetic associations involving TXNIP expression, TBI risk, NLRP3 and IL-1&#x3b2; expression. Collectively, these findings prioritize the TXNIP/NLRP3/CASP1 module as a computationally supported candidate mechanism through which HT may influence oxidative stress-inflammasome-immune coupling in TBI. This study provides an interpretable drug-target-pathway-phenotype framework and identifies TXNIP, NLRP3, and CASP1 as priority nodes for future experimental validation.

Artificial Intelligence↗
PubMed2026

Accurately Deciphering Tissue Heterogeneity From Spatial Multi-Modal and Multi-Omics With STransformer.

Advances in spatially resolved technologies enable the simultaneous acquisition of diverse data modalities within a tissue slice while preserving critical spatial context, which presents unprecedented opportunities to decipher intricate tissue heterogeneity. However, existing computational approaches lack the intrinsic flexibility to universally process both spatial multi-modal and multi-omics data. Here, we introduce STransformer, a unified deep learning framework designed to seamlessly accommodate a comprehensive landscape of spatial data. By simultaneously capturing short-range cellular interactions and tissue-wide semantic patterns, it extracts robust representations to accurately dissect complex tissue heterogeneity. Systematic evaluations across diverse species, tissue types, and data modalities highlight its profound versatility. For spatial multi-modal data, STransformer delineates intricate anatomical structures in the human cortex, uncovers pathological mechanisms in Alzheimer's disease, and characterizes dynamic spatiotemporal developmental trajectories during chicken cardiogenesis. Scaling to spatial multi-omics data, STransformer synergizes spatial transcriptomic and proteomic profiles to decipher intricate immune microenvironments within the human tonsil, and jointly analyzes spatial epigenomic and transcriptomic data to infer regulatory mechanisms in the mouse embryonic brain. Consequently, STransformer serves as a highly versatile and robust analytical framework for advancing our understanding of tissue heterogeneity and disease&#xa0;pathogenesis.

Multiomics↗
PubMed1996

Neural network simulation of visual inspection of graphs of single-subject interventions.

Judgments of the effectiveness of single-subject behavioral interventions are often based on visual examination of graphs of response data, but previous research indicates that such judgments are often unreliable and flawed. Here it is proposed that artificial neural networks could be developed to simulate the judgments of expert judges. A prototype of such a network was designed and trained in the present study, and its use in novel experiments matched the estimates of the expert whose judgments were simulated significantly better than did a prediction equation developed using a multiple regression approach.

Artificial Intelligence↗
PubMed2005

The loading problem for recursive neural networks.

The present work deals with one of the major and not yet completely understood topics of supervised connectionist models. Namely, it investigates the relationships between the difficulty of a given learning task and the chosen neural network architecture. These relationships have been investigated and nicely established for some interesting problems in the case of neural networks used for processing vectors and sequences, but only a few studies have dealt with loading problems involving graphical inputs. In this paper, we present sufficient conditions which guarantee the absence of local minima of the error function in the case of learning directed acyclic graphs with recursive neural networks. We introduce topological indices which can be directly calculated from the given training set and that allows us to design the neural architecture with local minima free error function. In particular, we conceive a reduction algorithm that involves both the information attached to the nodes and the topology, which enlarges significantly the class of the problems with unimodal error function previously proposed in the literature.

Algorithms↗
PubMed2004

Positive and negative circuits in discrete neural networks.

We study the relationships between the positive and negative circuits of the connection graph and the fixed points of discrete neural networks (DNNs). As main results, we give necessary conditions and sufficient conditions for the existence of fixed points in a DNN. Moreover, we exhibit an upper bound for the number of fixed points in terms of the structure and number of positive circuits in the connection graph. This allows the determination of the maximum capacity for storing vectors in DNNs as fixed points, depending on the architecture of the network.

Neural Networks, Computer↗
PubMed2005

On the relationship between deterministic and probabilistic directed Graphical models: from Bayesian networks to recursive neural networks.

Machine learning methods that can handle variable-size structured data such as sequences and graphs include Bayesian networks (BNs) and Recursive Neural Networks (RNNs). In both classes of models, the data is modeled using a set of observed and hidden variables associated with the nodes of a directed acyclic graph. In BNs, the conditional relationships between parent and child variables are probabilistic, whereas in RNNs they are deterministic and parameterized by neural networks. Here, we study the formal relationship between both classes of models and show that when the source nodes variables are observed, RNNs can be viewed as limits, both in distribution and probability, of BNs with local conditional distributions that have vanishing covariance matrices and converge to delta functions. Conditions for uniform convergence are also given together with an analysis of the behavior and exactness of Belief Propagation (BP) in 'deterministic' BNs. Implications for the design of mixed architectures and the corresponding inference algorithms are briefly discussed.

Bayes Theorem↗
PubMed2000

A signal-flow-graph approach to on-line gradient calculation.

A large class of nonlinear dynamic adaptive systems such as dynamic recurrent neural networks can be effectively represented by signal flow graphs (SFGs). By this method, complex systems are described as a general connection of many simple components, each of them implementing a simple one-input, one-output transformation, as in an electrical circuit. Even if graph representations are popular in the neural network community, they are often used for qualitative description rather than for rigorous representation and computational purposes. In this article, a method for both on-line and batch-backward gradient computation of a system output or cost function with respect to system parameters is derived by the SFG representation theory and its known properties. The system can be any causal, in general nonlinear and time-variant, dynamic system represented by an SFG, in particular any feedforward, time-delay, or recurrent neural network. In this work, we use discrete-time notation, but the same theory holds for the continuous-time case. The gradient is obtained in a straightforward way by the analysis of two SFGs, the original one and its adjoint (obtained from the first by simple transformations), without the complex chain rule expansions of derivatives usually employed. This method can be used for sensitivity analysis and for learning both off-line and on-line. On-line learning is particularly important since it is required by many real applications, such as digital signal processing, system identification and control, channel equalization, and predistortion.

Algorithms↗