PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “knowledge graphs”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 19 recordsLinked to original sources
PubMed2025

monarchr: an R package for querying biomedical knowledge graphs.

SUMMARY: Biomedical knowledge graphs (KGs) aggregate and provide a wealth of information, linking genes and their variants, diseases, phenotypes, and much more. While these data are available in raw and API-hosted form, to date, functionality for working with KGs in the R programming language has been limited. We introduce monarchr, a package for querying and manipulating KG data. Support for the expansive Monarch Initiative KG is built in, and monarchr can accommodate any KG in the Knowledge Graph eXchange (KGX) format. This tidy-inspired interface offers researchers an intuitive, iterative approach to querying and visualizing KG data. AVAILABILITY AND IMPLEMENTATION: Source code, documentation, and installation instructions are available at https://github.com/monarch-initiative/monarchr.

Software↗
PubMed2026

MetagenomicKG: a knowledge graph for metagenomic applications.

MOTIVATION: The sheer volume and variety of genomic content within microbial communities makes metagenomics a field rich in biomedical knowledge. To traverse these complex communities and their vast unknowns, metagenomic studies often depend on distinct reference databases, such as the Genome Taxonomy Database (GTDB), the Kyoto Encyclopedia of Genes and Genomes (KEGG), and the Bacterial and Viral Bioinformatics Resource Center (BV-BRC), for various analytical purposes. These databases are crucial for the genetic and functional annotation of microbial communities. Nevertheless, the inconsistent nomenclature or identifiers of these databases present challenges for effective integration, representation, and utilization. Knowledge graphs (KGs) offer an appropriate solution by organizing biological entities from different databases to standardized identifiers, allowing their interrelations to be captured into a cohesive network regardless of the naming conventions used in each source. The graph structure not only facilitates the unveiling of hidden patterns but also enriches our biological understanding with deeper insights. Despite KGs having shown potential in various biomedical fields, their application in metagenomics remains underexplored. RESULTS: We present MetagenomicKG, a novel knowledge graph specifically tailored for metagenomic analysis. MetagenomicKG integrates taxonomic, functional, and pathogenesis-related information on the human microbiome sourced from various databases, and further connects these with existing biomedical KGs to expand the biological network. Through various case studies involving the human microbiome, we demonstrate its utility in enabling hypothesis generation regarding the relationships between microbes and diseases, generating sample-specific graph embeddings, and providing robust pathogen prediction. CODE AVAILABILITY: The source code and technical details for constructing the MetagenomicKG and reproducing all analyses are available on GitHub at https://github.com/KoslickiLab/MetagenomicKG. The data used in this manuscript, including the pre-built files and use case input data, are archived on Zenodo with DOI: 10.5281/zenodo.17546861.

Metagenomics↗
PubMed2026

KG-Microbe: Building modular and scalable knowledge graphs for microbiome and microbial sciences.

BACKGROUND: The integration of many disparate forms of data is essential for understanding the microbial world and its interaction with the environment and human health. Doing so is particularly challenging in the context of microbe-host and microbe-microbe interactions that contribute to health or environmental outcomes. There are thousands of relevant microbial species, and millions of interactions among those microbes and with their environment or host. Integrated information (e.g., about host and microbial physiology, genetics, and metabolism) facilitates deeper understanding of complex mechanisms and helps interpret correlative results. RESULTS: The KG-Microbe construction framework is a novel approach to harmonizing bacterial and archaeal data in the form of a findable, accessible, interoperable, reusable and AI-ready knowledge graph (KG). Starting from a core KG with organismal traits, environments, and growth preferences and the integration of established ontologies, the framework generates a hierarchy of related KGs targeting specific use cases, including the human microbiome in the context of disease, or environmental microbiomes. The framework supports customizable taxa subsets representing communities or clades of interest. Evaluations of the KG-Microbe KGs through a series of competency questions demonstrate the accuracy and effectiveness of the data harmonization, and the utility of the resulting KGs in studies of inflammatory bowel disease and Parkinson's disease. Finally, the predictive and environmental capabilities of the KGs are demonstrated by predicting growth preferences using graph features. CONCLUSIONS: The KG-Microbe framework unifies microbial contexts in a single resource to support integrative analyses across biomedical, host, and environmental domains. KG-Microbe is a flexible, modular enabling technology for humans and machine learning methods to uncover candidate mechanistic explanations of microbial associations.

Microbiota↗
PubMed2025

Large Language Model and Knowledge Graph-Driven AJCC Staging of Prostate Cancer Using Pathology Reports.

Background/Objectives: To develop an automated American Joint Committee on Cancer (AJCC) staging system for radical prostatectomy pathology reports using large language model-based information extraction and knowledge graph validation. Methods: Pathology reports from 152 radical prostatectomy patients were used. Five additional parameters (Prostate-specific antigen (PSA) level, metastasis stage (M-stage), extraprostatic extension, seminal vesicle invasion, and perineural invasion) were extracted using GPT-4.1 with zero-shot prompting. A knowledge graph was constructed to model pathological relationships and implement rule-based AJCC staging with consistency validation. Information extraction performance was evaluated using a local open-source large language model (LLM) (Mistral-Small-3.2-24B-Instruct) across 16 parameters. The LLM-extracted information was integrated into the knowledge graph for automated AJCC staging classification and data consistency validation. The developed system was further validated using pathology reports from 88 radical prostatectomy patients in The Cancer Genome Atlas (TCGA) dataset. Results: Information extraction achieved an accuracy of 0.973 and an F1-score of 0.986 on the internal dataset, and 0.938 and 0.968, respectively, on external validation. AJCC staging classification showed macro-averaged F1-scores of 0.930 and 0.833 for the internal and external datasets, respectively. Knowledge graph-based validation detected data inconsistencies in 5 of 150 cases (3.3%). Conclusions: This study demonstrates the feasibility of automated AJCC staging through the integration of large language model information extraction and knowledge graph-based validation. The resulting system enables privacy-protected clinical decision support for cancer staging applications with extensibility to broader oncologic domains.

artificial intelligence↗
PubMed1997

[Toxoplasmosis diagnosis in pregnancy: computer assisted follow-up interpretation of serologic tests].

Primary infection with Toxoplasma gondii during pregnancy can result in fetal infection with serious sequelae for the unborn if not treated properly. Early diagnosis enables drug therapy and significantly reduces the risk of fetal disease. A systematic serological screening procedure was established in Austria in 1975 to detect primary toxoplasma infection as early as possible during pregnancy. Since the screening program is based solely on observation and interpretation of serological data, the question arises whether a knowledge-based system for automatic interpretation can achieve a sufficient interpretative accuracy for introduction to routine work. For this reason the system Toxopert-I was developed. The system is aimed at facilitating routine laboratory work, as well as assuring quality by setting standards for therapy. The required knowledge base was designed as a knowledge graph, each state representing a certain interpretation. One or more available serological test results cause the knowledge graph to change its current state. If all available test results are processed, the final state reached corresponds to the respective current interpretation for the patient. A retrospective analysis of 1000 pregnant women yielded a total diagnostic sensitivity and specificity of over 99% in comparison with the clinician's diagnosis which was used as the Gold Standard.

Animals↗
PubMed2025

BioMedGraphica: An All-in-One Platform for Joint Textual Biomedical Prior Knowledge and Numeric Graph Generation.

Multi-omic data analysis is essential for scientific discovery in precision medicine. However, translating statistical results of omic data analysis into novel scientific hypothesis remains a significant challenge. Human experts must manually review analysis results and generate new hypothesis based on extensive and inter-connected biomedical prior knowledge, which is subjective and not scalable. While large language models (LLMs) can accelerate the discovery, their reasoning improves when grounded in structured, auditable and comprehensive biomedical prior knowledge. Biomedical knowledge, however, is scattered across heterogeneous databases that use diverse and inconsistent nomenclature systems, making it difficult to integrate resources into a unified format for scalable analysis. This fragmentation limits the ability of AI systems to fully leverage biomedical data for scientific discovery. To address these challenges, we developed BioMedGraphica , an all-in-one platform that harmonizes fragmented biomedical resources by integrating 11 entity types and 30 relation types from 43 databases into a unified knowledge graph containing 2,306,921 entities and 27,232,091 relations. In addition, to the best of our knowledge, this is the first work to propose a novel Textual-Numeric Graph (TNG) data-structure for multi-omics data analysis. In TNG, textual information captures prior biological knowledge (e.g., transcription start sites, functions, mechanisms), while numeric values represent quantitative biomedical features, and the integrated relations can help uncover mechanisms. By bridging prior knowledge with user-specific data, TNG is a novel and ideal data-structure for the development of graph foundation models, with the potential to improve prediction performance and interpretability, while also augmenting LLMs by supplying graph-structured mechanistic context to strengthen reasoning. The details for BioMedGraphica code can be accessed by github link: https://github.com/FuhaiLiAiLab/BioMedGraphica and BioMedGraphica knowledge graph data can be downloaded from huggingface dataset: https://huggingface.co/datasets/FuhaiLiAiLab/BioMedGraphica.

biomedical knowledge graph↗
PubMed2026

BioMedGraphica: an all-in-one platform for joint textual biomedical prior knowledge and numeric graph generation.

MOTIVATION: Multiomics data analysis is essential for scientific discovery in precision medicine. However, translating analysis results of omics data analysis into novel scientific hypotheses remains a significant challenge. Human experts must manually review analysis results and generate new hypotheses based on extensive and interconnected biomedical prior knowledge, which is subjective and not scalable. While large language models can accelerate the discovery, their reasoning improves when grounded in structured, auditable, and comprehensive biomedical prior knowledge. However, biomedical knowledge is scattered across heterogeneous databases that use diverse and inconsistent nomenclature systems, making it difficult to integrate resources into a unified format for scalable analysis. This fragmentation limits the ability of artificial intelligence systems to fully leverage biomedical data for scientific discovery. RESULTS: We developed BioMedGraphica, a novel all-in-one platform that harmonizes fragmented biomedical resources by integrating 11 entity types and 30 relation types from 43 databases into a unified textual prior knowledge graph containing 2 306 921 entities and 27 232 091 relations. In addition, we present a novel textual-numeric graph (TNG) data structure concept, where textual information captures prior biological knowledge (e.g. transcription start sites, functions, mechanisms), numeric values represent quantitative biomedical features, and the integrated relations can help uncover mechanisms. By bridging prior knowledge with user-specific data, TNG is a novel and ideal data structure for developing novel graph analysis models. AVAILABILITY AND IMPLEMENTATION: The code is available at: https://github.com/FuhaiLiAiLab/BioMedGraphica and BioMedGraphica knowledge graph database can be downloaded from huggingface dataset: https://huggingface.co/datasets/FuhaiLiAiLab/BioMedGraphica.

Humans↗
PubMed2026

PheBee: A Graph-Aware System for Scalable, Traceable, and Semantic Phenotyping.

OBJECTIVES: Phenotype-driven workflows in clinical and translational research require standardized ontology-based representation, ontology-aware cohort discovery, and provenance inspection for each assertion. Existing approaches optimize either for semantic traversal or scalable batch analytics, but not both. We describe PheBee, a hybrid system that links semantic assertions to scalable evidence storage via a deterministic identifier, preserving provenance while supporting ontology-aware discovery at cohort scale. MATERIALS AND METHODS: PheBee represents phenotype assertions in a knowledge graph as ontology-linked nodes with clinical modifier context (e.g., negated, family history), and stores supporting evidence records in a scalable row-oriented evidence table for cohort-scale access. The two layers are connected by a deterministic identifier enabling stable joins across repeated ingestions without duplicating high-volume evidence in the graph. We evaluated PheBee using synthetic datasets designed to exercise end-to-end ingestion and query workflows. RESULTS: Functional evaluation validated hierarchical term expansion, qualifier-aware retrieval, duplicate-free assertion handling under re-ingestion, and privacy-conscious management of subjects shared across multiple research projects. At scale (10,000 subjects producing 12M evidence records) PheBee completed ingestion in ~30 minutes and responded to interactive queries within 6 seconds under concurrent load. DISCUSSION: PheBee exposes a unified API for ontology-aware cohort discovery with hierarchical term expansion, subject-centric retrieval of phenotypes and clinical modifiers, and evidence and provenance queries. Its data model aligns with GA4GH Phenopackets, facilitating interoperability with phenotype exchange standards. CONCLUSION: By combining ontology-aware semantics with scalable, provenance-bearing evidence storage, PheBee provides a practical open-source foundation for phenotype-driven research workflows that demand both semantic precision and cohort-scale traceability.

cohort studies↗
PubMed2025

The Data Distillery: A Graph Framework for Semantic Integration and Querying of Biomedical Data.

The Data Distillery Knowledge Graph (DDKG) is a framework for semantic integration and querying of biomedical data across domains. Built for the NIH Common Fund Data Ecosystem, it supports translational research by linking clinical and experimental datasets in a unified graph model. Clinical standards such as ICD-10, SNOMED, and DrugBank are integrated through UMLS, while genomics and basic science data are structured using ontologies and standards such as HPO, GENCODE, Ensembl, STRING, and ClinVar. The DDKG uses a property graph architecture based on the UBKG infrastructure and supports ontology-based ingestion, identifier normalization, and graph-native querying. The system is modular and can be extended with new datasets or schema modules. We demonstrate its utility for informatics queries across eight use cases, including regulatory variant analysis, tissue-specific expression, biomarker discovery, and cross-species variant prioritization. The DDKG is accessible via a public interface, a programmatic API, and downloadable builds for local use.

Journal Article↗
PubMed2025

Knowledge-enhanced protein subcellular localization prediction from 3D fluorescence microscope images.

MOTIVATION: Pinpointing the subcellular location of proteins is essential for studying protein function and related diseases. Advances in spatial proteomics have shown that automatic recognition of protein subcellular localization from images could highly facilitate protein translocation analysis and biomarker discovery, but existing machine-learning works have been mostly limited to processing 2D images. By contrast, 3D images have higher spatial resolution and allow researchers to observe cellular structures in their natural context, but currently, there are only a few studies of 3D image processing for protein distribution analysis due to the lack of data and complexity of modeling. RESULTS: We developed a knowledge-enhanced protein subcellular localization model, KE3DLoc, which could recognize distribution patterns in 3D fluorescence microscope images using deep learning methods. The model designs an image feature extraction module that incorporates information from 3D and 2D projected cells and implements asymmetric loss and confidence weights to address data imbalance and weak cell annotation issues. Besides, considering that the biological knowledge in the Gene Ontology (GO) database can provide valuable support for protein location understanding, the KE3DLoc model incorporates a novel knowledge enhancement module that optimizes the protein representation by related knowledge graphs derived from the GO. Since the image module and the knowledge module calculate features from different levels, KE3DLoc designs protein ID aggregation to enhance the consistency of protein features across different cells. Experimental results on three public datasets have demonstrated that the KE3DLoc significantly outperforms existing methods and provides valuable insights for spatial proteomics research. AVAILABILITY AND IMPLEMENTATION: All datasets and codes used in this study are available at GitHub: https://github.com/PRBioimages/KE3DLoc.

Microscopy, Fluorescence↗
PubMed2002

Speed of comprehension of visualized ordered sets.

The authors investigated the effects of visual properties of hierarchical graphs on speed of comprehension: planarity (crossing of lines), slopes (orientation of lines), and levels (adjustment of dots). In each of 4 experiments, 30 participants responded to interpretive questions that required comparisons among the elements of the graph. Knowledge provided to participants differed across Experiments 1a, 1b, and 1c; question demands varied in Experiment 2. Analysis of response latencies showed that crossing of lines is the most influential variable independent of orientation, dot adjustment, and question demands. Speed of comprehension decreased with increasing question demands. When question demands were high, orientation of lines also had an effect on speed of comprehension. Preliminary conclusions for the presentation of hierarchical graphs are drawn.

Adolescent↗
PubMed2025

Multi-Omics and Integrative Analytics in Natural Products Discovery.

Natural products (NPs) have long been an essential source of new bioactive compounds for drug discovery; however, traditional methods for screening and isolating these compounds can be slow and often yield diminishing returns. Fortunately, advanced multi-omics and computational approaches present powerful solutions to these challenges. This review highlights innovative methodologies that integrate metabolomics, genomics, transcriptomics, and proteomics with bioinformatics and analytical chemistry to accelerate NP discovery. For instance, untargeted metabolomics platforms like high-resolution liquid chromatography-tandem mass spectrometry (LC-MS/MS) and Global Natural Products Social (GNPS) molecular networking allow for comprehensive profiling of new compounds, while targeted isotope-labeling strategies enhance this process. Additionally, genome and metagenome mining tools such as antibiotics and secondary metabolite analysis shell (antiSMASH), Deep Biosynthetic Gene Cluster (DeepBGC), and Pipeline for Reconstructing Integrated Syntheses of Metabolites (PRISM) quickly identify biosynthetic gene clusters (BGCs) in both cultured and uncultured organisms, often using heterologous expression to validate products. Transcriptomic analyses, including RNA sequencing (RNA-seq), co-expression networks, and fluxomics, help clarify how pathways are regulated, while quantitative proteomics techniques like tandem mass tags/isobaric tags for relative and absolute quantitation (TMT/iTRAQ) and label-free methods, along with chemoproteomics approaches such as cellular thermal shift assay and thermal proteome profiling (TPP), uncover molecular targets and their mechanisms of action. This review also places significant emphasis on the role of artificial intelligence (AI) and machine learning (ML) in integrating multi-omics data, spanning activities from constructing gene-metabolite correlation networks to leveraging knowledge graphs and graph neural networks for data fusion and functional prediction. Finally, this review concludes by discussing the synergistic benefits of multi-omics for natural-product discovery, addressing current technical challenges, and exploring future directions toward high-throughput, intelligent data integration for next-generation NP research.

Biological Products↗
PubMed2002

Constructing knowledge. The role of graphs and tables in hard and soft psychology.

Because graphs provide a compact, rhetorically powerful way of representing research findings, recent theories of science have postulated their use as a distinguishing feature of science. Studies have shown that the use of graphs in journal articles correlates highly with the hardness of scientific fields, both across disciplines and across sub-fields of psychology. In contrast, the use of tables and inferential statistics in psychology is inversely related to subfield hardness, suggesting that the relationship between hardness and graph use is not attributable to differences in the use of quantitative data in subfields or their commitment to empiricism. Enhanced "graphicacy" among psychologists could contribute to the progress of psychological science by providing alternatives to significance testing and by facilitating communication across subfields.

Computer Graphics↗
PubMed1998

Review of biomedical knowledge and data representation with conceptual graphs.

The basis of conceptual graphs theory is an ontology of types of concepts. Concepts issued from the ontology are interlinked by semantic relationships and constitute canonical conceptual graphs. Canonical graphs may be combined to derive new conceptual graphs by means of formation rules. This formalism allows to separate knowledge representation into a conceptual level and a domain-dependent level, and enables to share and reuse a representation. This paper presents conceptual graph applications to biomedical data and concept representation, classification systems, information retrieval, and natural language understanding and processing. A discussion on the unifying role conceptual graphs theory plays in the implementation of knowledge-based systems is also presented.

Artificial Intelligence↗
PubMedDate not supplied

Generating graphs for visual analytics through interactive sketching.

We introduce an interactive graph generator, GreenSketch, designed to facilitate the creation of descriptive graphs required for different visual analytics tasks. The human-centric design approach of GreenSketch enables users to master the creation process without specific training or prior knowledge of graph model theory. The customized user interface encourages users to gain insight into the connection between the compact matrix representation and the topology of a graph layout when they sketch their graphs. Both the human-enforced and machine-generated randomnesses supported by GreenSketch provide the flexibility needed to address the uncertainty factor in many analytical tasks. This paper describes more than two dozen examples that cover a wide variety of graph creations from a single line of nodes to a real-life small-world network that describes a snapshot of telephone connections. While the discussion focuses mainly on the design of GreenSketch, we include a case study that applies the technology in a visual analytics environment and a usability study that evaluates the strengths and weaknesses of our design approach.

Algorithms↗
PubMed1993

Structuration and acquisition of medical knowledge. Using UMLS in the conceptual graph formalism.

The use of a taxonomy, such as the concept type lattice (CTL) of Conceptual Graphs, is a central structuring piece in a knowledge-based system. The knowledge it contains is constantly used by the system, and its structure provides a guide for the acquisition of other pieces of knowledge. We show how UMLS can be used as a knowledge resource to build a CTL and how the CTL can help the process of acquisition for other kinds of knowledge. We illustrate this method in the context of the MENELAS natural language understanding project.

Artificial Intelligence↗
PubMed2004

Bioelectric impedance vector distribution in peritoneal dialysis patients with different hydration status.

BACKGROUND: In continuous ambulatory peritoneal dialysis (CAPD), total body water (TBW) is estimated by functions of body weight, and by equations of bioelectric impedance analysis (BIA). These procedures may be biased with abnormal tissue hydration. We validated vector BIA (BIVA) patterns of hydration in CAPD patients, based on direct measurements of resistance (R) and reactance (Xc) (RXc graph) without knowledge of the body weight. METHODS: Cross-sectional study in 200 adult CAPD patients from two groups: 149 patients (77 males and 72 females) without edema (BMI 24.3 kg/m2), and 51 (29 males and 22 females) with pitting edema (BMI 24.6 kg/m2). Single frequency (50 kHz), whole-body impedance vector was measured with both empty and filled peritoneal cavity. Vector distribution was compared with that from 726 healthy subjects, 1116 hemodialysis patients, and 50 nephrotic patients, all with a same BMI. The performance of BIVA was compared with indications of four anthropometry and four conventional BIA equations for TBW. RESULTS: TBW estimates from anthropometry (Watson, Hume and Weyers, Chertow, and Johansson formulas) were misleading, indicating the same hydration in edema. TBW estimates from BIA equations indicated a 10% excess TBW in edema. BIVA were very sensitive to fluid overload, as both R (by 10%) and Xc (by 40%) were reduced in patients with edema (regardless of peritoneal filling). The vector distribution of individual CAPD patients without edema was superposable to that of the healthy, gender-specific, reference population (50%, 75%, and 95% tolerance ellipses, RXc graph) and close to the hemodialysis, presession distribution. Vectors from patients with edema were displaced downward on the RXc graph, out of the 75% ellipse (88% sensitivity and 87% specificity), and close to vectors from nephrotic patients. CONCLUSION: CAPD prescription would keep or bring vectors of patients back into the 75% reference ellipse (border for progression from latent to apparent overhydration across the lower pole) regardless of body weight. Whether CAPD patients with vector within the target ellipse have better outcome needs longitudinal evaluation.

Adolescent↗