Neuroanatomical domain of the foundational model of anatomy ontology

A computational ontology framework for the synthesis of multi-level pathology reports from brain MRI scans

BackgroundConvolutional neural network (CNN) based volumetry of MRI data can help differentiate Alzheimer's disease (AD) and the behavioral variant of frontotemporal dementia (bvFTD) as causes of cognitive decline and dementia. However, existing CNN-based MRI volumetry tools lack a structured hierarchical representation of brain anatomy, which would allow for aggregating regional pathological information and automated computational inference.ObjectiveDevelop a computational ontology pipeline for quantifying hierarchical pathological abnormalities and visualize summary charts for brain atrophy findings, aiding differential diagnosis.MethodsUsing FastSurfer, we segmented brain regions and measured volume and cortical thickness from MRI scans pooled across multiple cohorts (N = 3433; ADNI, AIBL, DELCODE, DESCRIBE, EDSD, and NIFD), including healthy controls, prodromal and clinical AD cases, and bvFTD cases. Employing the Web Ontology Language (OWL), we built a semantic model encoding hierarchical anatomical information. Additionally, we created summary visualizations based on sunburst plots for visual inspection of the information stored in the ontology.ResultsOur computational framework dynamically estimated and aggregated regional pathological deviations across different levels of neuroanatomy abstraction. The disease similarity index derived from the volumetric and cortical thickness deviations achieved an AUC of 0.88 for separating AD and bvFTD, which was also reflected by distinct atrophy profile visualizations.ConclusionsThe proposed automated pipeline facilitates visual comparison of atrophy profiles across various disease types and stages. It provides a generalizable computational framework for summarizing pathologic findings, potentially enhancing the physicians' ability to evaluate brain pathologies robustly and interpretably.

BioMedGraphica: An All-in-One Platform for Biomedical Prior Knowledge and Omic Signaling Graph Generation

Artificial intelligence (AI) is revolutionizing scientific discovery because of its super capability, following the neural scaling laws, to integrate and analyze large-scale datasets to mine knowledge. Foundation models, large language models (LLMs) and large vision models (LVMs), are among the most important foundations paving the way for general AI by pre-training on massive domain-specific datasets. Different from the well annotated, formatted and integrated large textual and image datasets for LLMs and LVMs, biomedical knowledge and datasets are fragmented with data scattered across publications and inconsistent databases that often use diverse nomenclature systems in the field of AI for Precision Health and Medicine (AI4PHM). These discrepancies, spanning different levels of biomedical organization from genes to clinical traits, present major challenges for data integration and alignment. To facilitate foundation AI model development and applications in AI4PHM, herein, we developed BioMedGraphica, an all-in-one platform and unified text-attributed knowledge graph (TAKG), consists of 3,131,788 entities and 56,817,063 relations, which are obtained from 11 distinct entity types and harmonizes 29 relations/edge types using data from 43 biomedical databases. All entities and relations are labeled a unique ID and associated with textual descriptions (textual features). Since covers most of research entities in AI4PHM, BioMedGraphica supports the zero-shot or few-shot knowledge discoveries via new relation prediction on the graph. Via a graphical user interface (GUI), researchers can access the knowledge graph with prior knowledge of target functional annotations, drugs, phenotypes and diseases (drug-protein-disease-phenotype), in the graph AI ready format. It also supports the generation of knowledge-multi-omic signaling graphs to facilitate the development and applications of novel AI models, like LLMs, graph AI, for AI4PHM science discovery, like discovering novel disease pathogenesis, signaling pathways, therapeutic targets, drugs and synergistic cocktails.

Disentangling associations between complex traits and cell types with seismic

Integrating single-cell RNA sequencing (scRNA-seq) with Genome-Wide Association Studies (GWAS) can help reveal GWAS-associated cell types, furthering our understanding of the cell-type-specific biological processes underlying complex traits and disease. However, current methods have technical limitations that hinder them from making systematic, scalable, interpretable disease-cell-type associations. In order to rapidly and accurately pinpoint associations, we develop a novel framework, seismic, which characterizes cell types using a new specificity score. We compare seismic with alternative methods across over 1,000 cell type characterizations at different granularities and 28 traits, demonstrating that seismic both corroborates findings and identifies trait-relevant cell groups which are not apparent through other methodologies. Furthermore, as part of the seismic framework, the specific genes driving cell type-trait associations can easily be accessed and analyzed, enabling further biological insights. The advantages of seismic are particularly salient in neurodegenerative diseases such as Parkinson's and Alzheimer's, where disease pathology has not only cell-specific manifestations, but also brain region-specific differences. Interestingly, a case study of Alzheimer's disease reveals the importance of considering GWAS endpoints, as studies relying on clinical diagnoses consistently identify microglial associations, while GWAS with a tau biomarker endpoint reveals neuronal associations. In general, seismic is a computationally efficient, powerful, and interpretable approach for identifying associations between complex traits and cell type-specific expression.

Drug-drug interactions prediction based on deep learning and knowledge graph: A review

Drug-drug interactions (DDIs) can produce unpredictable pharmacological effects and lead to adverse events that have the potential to cause irreversible damage to the organism. Traditional methods to detect DDIs through biological or pharmacological analysis are time-consuming and expensive, therefore, there is an urgent need to develop computational methods to effectively predict drug-drug interactions. Currently, deep learning and knowledge graph techniques which can effectively extract features of entities have been widely utilized to develop DDI prediction methods. In this research, we aim to systematically review DDI prediction researches applying deep learning and graph knowledge. The available biomedical data and public databases related to drugs are firstly summarized in this review. Then, we discuss the existing drug-drug interactions prediction methods which have utilized deep learning and knowledge graph techniques and group them into three main classes: deep learning-based methods, knowledge graph-based methods, and methods that combine deep learning with knowledge graph. We comprehensively analyze the commonly used drug related data and various DDI prediction methods, and compare these prediction methods on benchmark datasets. Finally, we briefly discuss the challenges related to drug-drug interactions prediction, including asymmetric DDIs prediction and high-order DDI prediction.

Loss of NMDA receptor function during development results in decreased KCC2 expression and increased neurons in the zebrafish forebrain

Developmental neurogenesis is a tightly regulated spatiotemporal process with its dysregulation implicated in neurodevelopmental disorders. NMDA receptors are glutamate-gated ion channels that are widely expressed in the early nervous system, yet their contribution to neurogenesis is poorly understood. Notably, a variety of mutations in genes encoding NMDA receptor subunits are associated with neurodevelopmental disorders. To rigorously define the role of NMDA receptors in developmental neurogenesis, we used a mutant zebrafish line ( grin1 -/- ) that lacks all NMDA receptors yet survives to 10 days post-fertilization, offering the opportunity to study post-embryonic neurodevelopment in the absence of NMDA receptors. Focusing on the forebrain, we find that these fish have a progressive supernumerary neuron phenotype confined to the telencephalon at the end of embryonic neurogenesis, but which extends to all forebrain regions during postembryonic neurogenesis. This enhanced neuron population does not arise directly from increased numbers or mitotic activity of radial glia cells, the principal neural stem cells. Rather, it stems from a lack of timely maturation of transit-amplifying neuroblasts into post-mitotic neurons, as indicated by a decrease in expression of the ontogenetically-expressed chloride transporter, KCC2. Pharmacological blockade with MK-801 recapitulates the grin1 -/- supernumerary neuron phenotype, indicating a requirement for ionotropic signaling. Thus, NMDA receptors are required for suppression of indirect, transit amplifying cell-driven neurogenesis by promoting maturational termination of mitosis. Loss of suppression results in neuronal overpopulation that can fundamentally change brain circuitry and may be a key factor in pathogenesis of neurodevelopmental disorders caused by NMDA receptor dysfunction.

An ontology-driven framework for knowledge representation of digital extortion attacks

With the COVID-19 pandemic and the growing influence of the Internet in critical sectors of industry and society, cyberattacks have not only not declined, but have risen sharply. In the meantime, ransomware is at the forefront of the most devastating threats that have launched the lucrative illegal business. Due to the proliferation and variety of ransomware forays, there is a need for a new theory of categories. The intricacy and multiplicity of components involved in digital extortions entails the construction of a knowledge representation system that is able to organize large volumes of information from heterogeneous sources in a formal structured format and infer new knowledge from it. This paper suggests and develops a dedicated ontology of digital blackmails, called Rantology, with a particular focus on ransomware assaults. The logic coded in this ontology allows to assess the maliciousness of programs based on various factors, including called API functions and their behaviors. The proposed framework can be used to facilitate interoperability between cybersecurity experts and knowledge-based systems, and identify sensitive points for surveillance. The evaluation results based on several criteria confirm the adequacy of the suggested ontology in terms of clarity, modularity, consistency, coverage and inheritance richness.

Extending and using anatomical vocabularies in the stimulating peripheral activity to relieve conditions project

The stimulating peripheral activity to relieve conditions (SPARC) program is a US National Institutes of Health-funded effort to improve our understanding of the neural circuitry of the autonomic nervous system (ANS) in support of bioelectronic medicine. As part of this effort, the SPARC project is generating multi-species, multimodal data, models, simulations, and anatomical maps supported by a comprehensive knowledge base of autonomic circuitry. To facilitate the organization of and integration across multi-faceted SPARC data and models, SPARC is implementing the findable, accessible, interoperable, and reusable (FAIR) data principles to ensure that all SPARC products are findable, accessible, interoperable, and reusable. We are therefore annotating and describing all products with a common FAIR vocabulary. The SPARC Vocabulary is built from a set of community ontologies covering major domains relevant to SPARC, including anatomy, physiology, experimental techniques, and molecules. The SPARC Vocabulary is incorporated into tools researchers use to segment and annotate their data, facilitating the application of these ontologies for annotation of research data. However, since investigators perform deep annotations on experimental data, not all terms and relationships are available in community ontologies. We therefore implemented a term management and vocabulary extension pipeline where SPARC researchers may extend the SPARC Vocabulary using InterLex, an online vocabulary management system. To ensure the quality of contributed terms, we have set up a curated term request and review pipeline specifically for anatomical terms involving expert review. Accepted terms are added to the SPARC Vocabulary and, when appropriate, contributed back to community ontologies to enhance ANS coverage. Here, we provide an overview of the SPARC Vocabulary, the infrastructure and process for implementing the term management and review pipeline. In an analysis of >300 anatomical contributed terms, the majority represented composite terms that necessitated combining terms within and across existing ontologies. Although these terms are not good candidates for community ontologies, they can be linked to structures contained within these ontologies. We conclude that the term request pipeline serves as a useful adjunct to community ontologies for annotating experimental data and increases the FAIRness of SPARC data.

MRI-based Parcellation and Morphometry of the Individual Rhesus Monkey Brain: the macaque Harvard-Oxford Atlas (mHOA), a translational system referencing a standardized ontology

Investigations of the rhesus monkey (Macaca mulatta) brain have shed light on the function and organization of the primate brain at a scale and resolution not yet possible in humans. A cornerstone of the linkage between non-human primate and human studies of the brain is magnetic resonance imaging, which allows for an association to be made between the detailed structural and physiological analysis of the non-human primate and that of the human brain. To further this end, we present a novel parcellation method and system for the rhesus monkey brain, referred to as the macaque Harvard-Oxford Atlas (mHOA), which is based on the human Harvard-Oxford Atlas (HOA) and grounded in an ontological and taxonomic framework. Consistent anatomical features were used to delimit and parcellate brain regions in the macaque, which were then categorized according to functional systems. This system of parcellation will be expanded with advances in technology and, like the HOA, will provide a framework upon which the results from other experimental studies (e.g., functional magnetic resonance imaging (fMRI), physiology, connectivity, graph theory) can be interpreted.

Bayes Performance of Batch Data Mining Based on Functional Dependencies

Batch processes and phenomena in traffic video data processing, such as traffic video image processing and intelligent transportation, are commonly used. The application of batch processing can increase the efficiency of resource conservation. However, owing to limited research on traffic video data processing conditions, batch processing activities in this area remain minimally examined. By employing database functional dependency mining, we developed in this study a workflow system. Meanwhile, the Bayesian network is a focus area of data mining. It provides an intuitive means for users to comply with causality expression approaches. Moreover, graph theory is also used in data mining area. In this study, the proposed approach depends on relational database functions to remove redundant attributes, reduce interference, and select a property order. The restoration of selective hidden naive Bayesian (SHNB) affects this property order when it is used only once. With consideration of the hidden naive Bayes (HNB) influence, rather than using one pair of HNB, it is introduced twice. We additionally designed and implemented mining dependencies from a batch traffic video processing log for data execution algorithms.

From Matrices to Knowledge: Using Semantic Networks to Annotate the Connectome

The connectome is regarded as the key to brain function in health and disease. Structural and functional neuroimaging enables us to measure brain connectivity in the living human brain. The field of connectomics describes the connectome as a mathematical graph with its connection strengths being represented by connectivity matrices. Graph theory algorithms are used to assess the integrity of the graph as a whole and to reveal brain network biomarkers for brain diseases; however, the faulty wiring of single connections or subnetworks as the structural correlate for neurological or mental diseases remains elusive. We describe a novel approach to represent the knowledge of human brain connectivity by a semantic network – a formalism frequently used in knowledge management to describe the semantic relations between objects. In our novel approach, objects are brain areas and connectivity is modeled as semantic relations among them. The semantic network turns the graph of the connectome into an explicit knowledge base about which brain areas are interconnected. Moreover, this approach can semantically enrich the measured connectivity of an individual subject by the semantic context from ontologies, brain atlases and molecular biological databases. Integrating all measurements and facts into one unified feature space enables cross-modal comparisons and analyses. We used a query mechanism for semantic networks to extract functional, structural and transcriptome networks. We found that in general higher structural and functional connectivity go along with a lower differential gene expression among connected brain areas; however, subcortical motor areas and limbic structures turned out to have a localized high differential gene expression while being strongly connected. In an additional explorative use case, we could show a localized high availability of fkbp5, gmeb1, and gmeb2 genes at a connection hub of temporo-limbic brain networks. Fkbp5 is known for having a role in stress-related psychiatric disorders, while gmeb1 and gmeb2 encode for modulator proteins of the glucocorticoid receptor, a key receptor in the hormonal stress system. Semantic networks tremendously ease working with multimodal neuroimaging and neurogenetics data and may reveal relevant coincidences between transcriptome and connectome networks.

Neuroinformatics Software Applications Supporting Electronic Data Capture, Management, and Sharing for the Neuroimaging Community

Accelerating insight into the relation between brain and behavior entails conducting small and large-scale research endeavors that lead to reproducible results. Consensus is emerging between funding agencies, publishers, and the research community that data sharing is a fundamental requirement to ensure all such endeavors foster data reuse and fuel reproducible discoveries. Funding agency and publisher mandates to share data are bolstered by a growing number of data sharing efforts that demonstrate how information technologies can enable meaningful data reuse. Neuroinformatics evaluates scientific needs and develops solutions to facilitate the use of data across the cognitive and neurosciences. For example, electronic data capture and management tools designed to facilitate human neurocognitive research can decrease the setup time of studies, improve quality control, and streamline the process of harmonizing, curating, and sharing data across data repositories. In this article we outline the advantages and disadvantages of adopting software applications that support these features by reviewing the tools available and then presenting two contrasting neuroimaging study scenarios in the context of conducting a cross-sectional and a multisite longitudinal study.

Knowledge Representation and Management. From Ontology to Annotation. Findings from the Yearbook 2015 Section on Knowledge Representation and Management

OBJECTIVE

To summarize the best papers in the field of Knowledge Representation and Management (KRM).

METHODS

A comprehensive review of medical informatics literature was performed to select some of the most interesting papers of KRM published in 2014.

RESULTS

Four articles were selected, two focused on annotation and information retrieval using an ontology. The two others focused mainly on ontologies, one dealing with the usage of a temporal ontology in order to analyze the content of narrative document, one describing a methodology for building multilingual ontologies.

CONCLUSION

Semantic models began to show their efficiency, coupled with annotation tools.

Terminology development towards harmonizing multiple clinical neuroimaging research repositories

Data sharing and mediation across disparate neuroimaging repositories requires extensive effort to ensure that the different domains of data types are referred to by commonly agreed upon terms. Within the SchizConnect project, which enables querying across decentralized databases of neuroimaging, clinical, and cognitive data from various studies of schizophrenia, we developed a model for each data domain, identified common usable terms that could be agreed upon across the repositories, and linked them to standard ontological terms where possible. We had the goal of facilitating both the current user experience in querying and future automated computations and reasoning regarding the data. We found that existing terminologies are incomplete for these purposes, even with the history of neuroimaging data sharing in the field; and we provide a model for efforts focused on querying multiple clinical neuroimaging repositories.

Ontology-based approach for in vivo human connectomics: the medial Brodmann area 6 case study

Different non-invasive neuroimaging modalities and multi-level analysis of human connectomics datasets yield a great amount of heterogeneous data which are hard to integrate into an unified representation. Biomedical ontologies can provide a suitable integrative framework for domain knowledge as well as a tool to facilitate information retrieval, data sharing and data comparisons across scales, modalities and species. Especially, it is urgently needed to fill the gap between neurobiology and in vivo human connectomics in order to better take into account the reality highlighted in Magnetic Resonance Imaging (MRI) and relate it to existing brain knowledge. The aim of this study was to create a neuroanatomical ontology, called “Human Connectomics Ontology” (HCO), in order to represent macroscopic gray matter regions connected with fiber bundles assessed by diffusion tractography and to annotate MRI connectomics datasets acquired in the living human brain. First a neuroanatomical “view” called NEURO-DL-FMA was extracted from the reference ontology Foundational Model of Anatomy (FMA) in order to construct a gross anatomy ontology of the brain. HCO extends NEURO-DL-FMA by introducing entities (such as “MR_Node” and “MR_Route”) and object properties (such as “tracto_connects”) pertaining to MR connectivity. The Web Ontology Language Description Logics (OWL DL) formalism was used in order to enable reasoning with common reasoning engines. Moreover, an experimental work was achieved in order to demonstrate how the HCO could be effectively used to address complex queries concerning in vivo MRI connectomics datasets. Indeed, neuroimaging datasets of five healthy subjects were annotated with terms of the HCO and a multi-level analysis of the connectivity patterns assessed by diffusion tractography of the right medial Brodmann Area 6 was achieved using a set of queries. This approach can facilitate comparison of data across scales, modalities and species.

Informatic system for a global tissue-fluid biorepository with a graph theory-oriented graphical user interface

The Richard Floor Biorepository supports collaborative studies of extracellular vesicles (EVs) found in human fluids and tissue specimens. The current emphasis is on biomarkers for central nervous system neoplasms but its structure may serve as a template for collaborative EV translational studies in other fields. The informatic system provides specimen inventory tracking with bar codes assigned to specimens and containers and projects, is hosted on globalized cloud computing resources, and embeds a suite of shared documents, calendars, and video-conferencing features. Clinical data are recorded in relation to molecular EV attributes and may be tagged with terms drawn from a network of externally maintained ontologies thus offering expansion of the system as the field matures. We fashioned the graphical user interface (GUI) around a web-based data visualization package. This system is now in an early stage of deployment, mainly focused on specimen tracking and clinical, laboratory, and imaging data capture in support of studies to optimize detection and analysis of brain tumour-specific mutations. It currently includes 4,392 specimens drawn from 611 subjects, the majority with brain tumours. As EV science evolves, we plan biorepository changes which may reflect multi-institutional collaborations, proteomic interfaces, additional biofluids, changes in operating procedures and kits for specimen handling, novel procedures for detection of tumour-specific EVs, and for RNA extraction and changes in the taxonomy of EVs. We have used an ontology-driven data model and web-based architecture with a graph theory-driven GUI to accommodate and stimulate the semantic web of EV science.