Challenges in the construction of knowledge bases for human microbiome-disease associations

Microbiome miracles and their pioneering advances and future frontiers in cardiovascular disease

Cardiovascular diseases (CVDs) represent a significant global health challenge, underscoring the need for innovative approaches to prevention and treatment. Recent years have seen a surge in interest in unraveling the complex relationship between the gut microbiome and cardiovascular health. This article delves into current research on the composition, diversity, and impact of the gut microbiome on CVD development. Recent advancements have elucidated the profound influence of the gut microbiome on disease progression, particularly through key mediators like Trimethylamine-N-oxide (TMAO) and other microbial metabolites. Understanding these mechanisms reveals promising therapeutic targets, including interventions aimed at modulating the gut microbiome's interaction with the immune system and its contribution to endothelial dysfunction. Harnessing this understanding, personalized medicine strategies tailored to individuals' gut microbiome profiles offer innovative avenues for reducing cardiovascular risk. As research in this field continues to evolve, there is vast potential for transformative advancements in cardiovascular medicine, paving the way for precision prevention and treatment strategies to address this global health challenge.

The gut microbiome as a modulator of arterial function and age-related arterial dysfunction

The arterial system is integral to the proper function of all other organs and tissues. Arterial function is impaired with aging, and arterial dysfunction contributes to the development of numerous age-related diseases, including cardiovascular diseases. The gut microbiome has emerged as an important regulator of both normal host physiological function and impairments in function with aging. The purpose of this review is to summarize more recently published literature demonstrating the role of the gut microbiome in supporting normal arterial development and function and in modulating arterial dysfunction with aging in the absence of overt disease. The gut microbiome can be altered due to a variety of exposures, including physiological aging processes. We explore mechanisms by which the gut microbiome may contribute to age-related arterial dysfunction, with a focus on changes in various gut microbiome-related compounds in circulation. In addition, we discuss how modulating circulating levels of these compounds may be a viable therapeutic approach for improving artery function with aging. Finally, we identify and discuss various experimental considerations and research gaps/areas of future research.

Hypothesizing mechanistic links between microbes and disease using knowledge graphs

Knowledge graphs have found broad biomedical applications, providing useful representations of complex knowledge. Although plentiful evidence exists linking the gut microbiome to disease, mechanistic understanding of those relationships remains generally elusive. Here we demonstrate the potential of knowledge graphs to hypothesize plausible mechanistic accounts of host-microbe interactions in disease. To do so, we constructed a knowledge graph of linked microbes, genes and metabolites called MGMLink. Using a semantically constrained shortest path search through the graph and a novel path prioritization methodology based on cosine similarity, we show that this knowledge supports inference of mechanistic hypotheses that explain observed relationships between microbes and disease phenotypes. We discuss specific applications of this methodology in inflammatory bowel disease and Parkinson's disease. This approach enables mechanistic hypotheses surrounding the complex interactions between gut microbes and disease to be generated in a scalable and comprehensive manner.

Leveraging pre-trained language models for mining microbiome-disease relationships

BACKGROUND

The growing recognition of the microbiome's impact on human health and well-being has prompted extensive research into discovering the links between microbiome dysbiosis and disease (healthy) states. However, this valuable information is scattered in unstructured form within biomedical literature. The structured extraction and qualification of microbe-disease interactions are important. In parallel, recent advancements in deep-learning-based natural language processing algorithms have revolutionized language-related tasks such as ours. This study aims to leverage state-of-the-art deep-learning language models to extract microbe-disease relationships from biomedical literature.

RESULTS

In this study, we first evaluate multiple pre-trained large language models within a zero-shot or few-shot learning context. In this setting, the models performed poorly out of the box, emphasizing the need for domain-specific fine-tuning of these language models. Subsequently, we fine-tune multiple language models (specifically, GPT-3, BioGPT, BioMedLM, BERT, BioMegatron, PubMedBERT, BioClinicalBERT, and BioLinkBERT) using labeled training data and evaluate their performance. Our experimental results demonstrate the state-of-the-art performance of these fine-tuned models ( specifically GPT-3, BioMedLM, and BioLinkBERT), achieving an average F1 score, precision, and recall of over [Formula: see text] compared to the previous best of  0.74.

CONCLUSION

Overall, this study establishes that pre-trained language models excel as transfer learners when fine-tuned with domain and problem-specific data, enabling them to achieve state-of-the-art results even with limited training data for extracting microbiome-disease interactions from scientific publications.

KG-Hub-building and exchanging biological knowledge graphs

MOTIVATION

Knowledge graphs (KGs) are a powerful approach for integrating heterogeneous data and making inferences in biology and many other domains, but a coherent solution for constructing, exchanging, and facilitating the downstream use of KGs is lacking.

RESULTS

Here we present KG-Hub, a platform that enables standardized construction, exchange, and reuse of KGs. Features include a simple, modular extract-transform-load pattern for producing graphs compliant with Biolink Model (a high-level data model for standardizing biological data), easy integration of any OBO (Open Biological and Biomedical Ontologies) ontology, cached downloads of upstream data sources, versioned and automatically updated builds with stable URLs, web-browsable storage of KG artifacts on cloud infrastructure, and easy reuse of transformed subgraphs across projects. Current KG-Hub projects span use cases including COVID-19 research, drug repurposing, microbial-environmental interactions, and rare disease research. KG-Hub is equipped with tooling to easily analyze and manipulate KGs. KG-Hub is also tightly integrated with graph machine learning (ML) tools which allow automated graph ML, including node embeddings and training of models for link prediction and node classification.

AVAILABILITY AND IMPLEMENTATION

https://kghub.org.

Dietary Efficacy Evaluation by Applying a Prediction Model Using Clinical Fecal Microbiome Data of Colorectal Disease to a Controlled Animal Model from an Obesity Perspective

Obesity associated with a Western diet such as a high-fat diet (HFD) is a known risk factor for inflammatory bowel disease (IBD) and colorectal cancer (CRC). In this study, we aimed to develop fecal microbiome data-based deep learning algorithms for the risk assessment of colorectal diseases. The effects of a HFD and a candidate food (Nypa fruticans, NF) on IBD and CRC risk reduction were also evaluated. Fecal microbiome data were obtained from 109 IBD patients, 111 CRC patients, and 395 healthy control (HC) subjects by 16S rDNA amplicon sequencing. IBD and CRC risk assessment prediction models were then constructed by deep learning algorithms. Dietary effects were evaluated based on fecal microbiome data from rats fed on a regular chow diet (RCD), HFD, and HFD plus ethanol extracts or water extracts of NF. There were significant differences in taxa when IBD and CRC were compared with HC. The diagnostic performance (area under curve, AUC) of the deep learning algorithm was 0.84 for IBD and 0.80 for CRC prediction. Based on the rat fecal microbiome data, IBD and CRC risks were increased in HFD-fed rats versus RCD-fed rats. Interestingly, in the HFD-induced obesity model, the IBD and CRC risk scores were significantly lowered by the administration of ethanol extracts of NF, but not by the administration of water extracts of NF. In conclusion, changes in the fecal microbiome of obesity by Western diet could be important risk factors for the development of IBD and CRC. The risk prediction model developed in this study could be used to evaluate dietary efficacy.

Machine-learning algorithms for asthma, COPD, and lung cancer risk assessment using circulating microbial extracellular vesicle data and their application to assess dietary effects

Although mounting evidence suggests that the microbiome has a tremendous influence on intractable disease, the relationship between circulating microbial extracellular vesicles (EVs) and respiratory disease remains unexplored. Here, we developed predictive diagnostic models for COPD, asthma, and lung cancer by applying machine learning to microbial EV metagenomes isolated from patient serum and coded by their accumulated taxonomic hierarchy. All models demonstrated high predictive strength with mean AUC values ranging from 0.93 to 0.99 with various important features at the genus and phylum levels. Application of the clinical models in mice showed that various foods reduced high-fat diet-associated asthma and lung cancer risk, while COPD was minimally affected. In conclusion, this study offers a novel methodology for respiratory disease prediction and highlights the utility of serum microbial EVs as data-rich features for noninvasive diagnosis.

Artificial intelligence (AI) has enabled researchers to intercept microbial messages bearing clinically useful information about of a variety of respiratory disorders. The organisms that comprise our microbiome communicate via the release of tiny, biomolecule-laden membrane bubbles called ‘extracellular vesicles’ (EVs) into the bloodstream. EVs are also influenced by human disease. South Korean researchers led by Yoon-Keun Kim of MD Healthcare, Seoul, and Young-Koo Jee of Dankook University College of Medicine, Cheonan, have used an AI algorithm to assemble EV-based profiles that can discriminate between healthy people and those with conditions like asthma or lung cancer. Their analysis of 1727 patient serum samples revealed microbial signatures that enabled accurate diagnosis of several respiratory disorders. Preliminary experiments in mice suggest that certain dietary changes could help shift the microbiome of high-risk individuals towards a healthier profile.

The Emerging Role of the Gut Microbiome in Cardiovascular Disease: Current Knowledge and Perspectives

The collection of normally non-pathogenic microorganisms that mainly inhabit our gut lumen shapes our health in many ways. Structural and functional perturbations in the gut microbial pool, known as “dysbiosis”, have been proven to play a vital role in the pathophysiology of several diseases, including cardiovascular disease (CVD). Although therapeutic regimes are available to treat this group of diseases, they have long been the main cause of mortality and morbidity worldwide. While age, sex, genetics, diet, tobacco use, and alcohol consumption are major contributors (World Health Organization, 2018), they cannot explain all of the consequences of CVD. In addition to the abovementioned traditional risk factors, the constant search for novel preventative and curative tools has shed light on the involvement of gut bacteria and their metabolites in the pathogenesis of CVD. In this narrative review, we will discuss the established interconnections between the gut microbiota and CVD, as well as the plausible therapeutic perspectives.

Reporting guidelines for human microbiome research: the STORMS checklist

The particularly interdisciplinary nature of human microbiome research makes the organization and reporting of results spanning epidemiology, biology, bioinformatics, translational medicine and statistics a challenge. Commonly used reporting guidelines for observational or genetic epidemiology studies lack key features specific to microbiome studies. Therefore, a multidisciplinary group of microbiome epidemiology researchers adapted guidelines for observational and genetic studies to culture-independent human microbiome studies, and also developed new reporting elements for laboratory, bioinformatics and statistical analyses tailored to microbiome studies. The resulting tool, called 'Strengthening The Organization and Reporting of Microbiome Studies' (STORMS), is composed of a 17-item checklist organized into six sections that correspond to the typical sections of a scientific publication, presented as an editable table for inclusion in supplementary materials. The STORMS checklist provides guidance for concise and complete reporting of microbiome studies that will facilitate manuscript preparation, peer review, and reader comprehension of publications and comparative analysis of published results.

Can manipulation of gut microbiota really be transformed into an intervention strategy for cardiovascular disease management?

Recent advancement in manipulation techniques of gut microbiota either ex vivo or in situ has broadened its plausible applicability for treating various diseases including cardiovascular disease. Several reports suggested that altering gut microbiota composition is an effective way to deal with issues associated with managing cardiovascular diseases. However, actual translation of gut microbiota manipulation-based techniques into cardiovascular-therapeutic approach is still questionable. This review summarized the evidence on challenges, opportunities, recent development, and future prospects of gut microbiota manipulation for targeting cardiovascular diseases. Initially, issues associated with current cardiovascular diseases treatment strategy, association of gut microbiota with cardiovascular disease, and its influence on cardiovascular drugs were discussed, followed by applicability of gut microbiota manipulation as a cardiovascular disease intervention strategy along with its challenges and future prospects. Despite the fact that the gut microbiota is rugged, interventions like probiotics, prebiotics, synbiotics, fecal microbiota transplantation, fecal virome transplantation, antibiotics, diet changes, and exercises could manipulate it. Advanced techniques like administration of engineered bacteriophages and bacteria could also be employed. Intensive exploration revealed that if sufficiently controlled approach and proper monitoring were applied, gut microbiota could provide a compelling answer for cardiovascular therapy.

Gastritis: The clinico-pathological spectrum

The inflammatory spectrum of gastric diseases includes different clinico-pathological entities, the etiology of which was recently established in the international Kyoto classification. A diagnosis of gastritis combines the information resulting form the gross examination (endoscopy) and histology (microscopy). It is important to consider the anatomical/functional heterogeneity of the gastric mucosa when obtaining representative mucosal biopsy samples. Gastritis includes self-limiting and non-self-limiting (long-standing) inflammatory diseases, and the latter are epidemiologically, biologically and clinically linked to the onset of gastric cancer (i.e. "inflammation-associated cancer"). Different biological models of inflammation-associated gastric oncogenesis have been proposed. Helicobacter pylori (H. pylori) gastritis is the most prevalent worldwide, and H. pylori is classified as a first-class carcinogen. On these bases, eradicating H. pylori is mandatory for the primary prevention of gastric cancer. Non-self-limiting gastritis may also be triggered by the immune-mediated destruction of gastric parietal cells, resulting in autoimmune gastritis. In both H. pylori-related and autoimmune gastritis, the non-self-limiting inflammation results in atrophy of the gastric mucosa, which is the main factor promoting gastric cancer. Long-term follow-up studies consistently demonstrate the prognostic impact of the histological staging of gastritis in gastric cancer secondary prevention strategies.

Nasopharyngeal microbiome composition associated with Streptococcus pneumoniae colonization suggests a protective role of Corynebacterium in young children

Streptococcus pneumoniae (Spn) is a leading respiratory tract pathogen that colonizes the nasopharynx (NP) through adhesion to epithelial cells and immune evasion. Spn actively interacts with other microbiota in NP but the nature of these interactions are incompletely understood. Using 16S rRNA gene sequencing, we analyzed the microbiota composition in the NP of children with or without Spn colonization. 96 children were included in the study cohort. 74 NP samples were analyzed when children were 6 months old and 85 NP samples were analyzed when children were 12 months old. We found several genera that correlated negatively or positively with Spn colonization, and some of these correlations appeared to be influenced by daycare attendance or other confounding factors such as upper respiratory infection (URI) or Moraxella colonization. Among these genera, Corynebacterium showed a consistent inverse relationship with Spn colonization with little influence by daycare attendance or other factors. We isolated Corynebacterium propinquum and C. pseudodiphtheriticum and found that both inhibited the growth of Spn serotype 22F strain in vitro.

Nasopharyngeal microbiome analyses in otitis-prone and otitis-free children

OBJECTIVES

About 10-15% children develop frequent acute otitis media (AOM) confirmed by tympanocentesis. These children are designated sOP (stringently defined otitis-prone) because all AOM episodes have been microbiologically confirmed. The cause of otitis-proneness in sOP children is multi-factorial, including frequent otopathogen nasopharyngeal (NP) colonization and deficiency in innate and adaptive immune responses. A largely unexplored contributor to otitis proneness is NP microbiome composition. Since the microbiome modulates otopathogen NP colonization and immune responses, we hypothesized that the NP microbiome composition in sOP children might be dysregulated.

METHODS

We performed 16S rRNA sequencing to analyze microbiome composition in 157 NP samples from 28 sOP and 68 AOM-free children when they were 6 months or 12 months old and healthy. Bioinformatic approaches were employed to examine the composition difference between the two populations and its correlation with changes in levels of inflammatory cytokines.

RESULTS

A different global microbiome profile and reduced alpha diversity was observed in the NP microbiome of sOP children when 6 months old, compared with that from AOM-free children of the same age. This difference was resolved when groups were compared at 12 months old. We found 4 bacterial genera-Bacillus, Veillonella, Gemella, and Prevotella-correlated with higher levels of pro-inflammatory cytokines in the NP. Those 4 bacterial genera were in lower abundance in sOP compared to AOM-free children.

CONCLUSION

Dysbiosis occurs in the NP microbiome of sOP children at an early age even when they were healthy. This dysbiosis correlates with a lower inflammatory state in the NP of these children.

GutBalance: a server for the human gut microbiome-based disease prediction and biomarker discovery with compositionality addressed

The compositionality of the microbiome data is well-known but often neglected. The compositional transformation pertains to the supervised learning of microbiome data and is a critical step that decides the performance and reliability of the disease classifiers. We value the excellent performance of the distal discriminative balance analysis (DBA) method, which selects distal balances of pairs and trios of bacteria, in addressing the classification of high-dimensional microbiome data. By applying this method to the species-level abundances of all the disease phenotypes in the GMrepo database, we build a balance-based model repository for the classification of human gut microbiome-related diseases. The model repository supports the prediction of disease risks for new sample(s). More importantly, we highlight the concept of balance-disease associations rather than the conventional microbe-disease associations and develop the human Gut Balance-Disease Association Database (GBDAD). Each predictable balance for each disease model indicates a potential biomarker-disease relationship and can be interpreted as a bacteria ratio positively or negatively correlated with the disease. Furthermore, by linking the balance-disease associations to the evidenced microbe-disease associations in MicroPhenoDB, we surprisingly found that most species-disease associations inferred from the shotgun metagenomic datasets can be validated by external evidence beyond MicroPhenoDB. The balance-based species-disease association inference will accelerate the generation of new microbe-disease association hypotheses in gastrointestinal microecology research and clinical trials. The model repository and the GBDAD database are deployed on the GutBalance server, which supports interactive visualization and systematic interrogation of the disease models, disease-related balances and disease-related species of interest.

Peryton: a manual collection of experimentally supported microbe-disease associations

We present Peryton (https://dianalab.e-ce.uth.gr/peryton/), a database of experimentally supported microbe-disease associations. Its first version constitutes a novel resource hosting more than 7900 entries linking 43 diseases with 1396 microorganisms. Peryton's content is exclusively sustained by manual curation of biomedical articles. Diseases and microorganisms are provided in a systematic, standardized manner using reference resources to create database dictionaries. Information about the experimental design, study cohorts and the applied high- or low-throughput techniques is meticulously annotated and catered to users. Several functionalities are provided to enhance user experience and enable ingenious use of Peryton. One or more microorganisms and/or diseases can be queried at the same time. Advanced filtering options and direct text-based filtering of results enable refinement of returned information and the conducting of tailored queries suitable to different research questions. Peryton also provides interactive visualizations to effectively capture different aspects of its content and results can be directly downloaded for local storage and downstream analyses. Peryton will serve as a valuable source, enabling scientists of microbe-related disease fields to form novel hypotheses but, equally importantly, to assist in cross-validation of findings.

Current Concepts, Opportunities, and Challenges of Gut Microbiome-Based Personalized Medicine in Nonalcoholic Fatty Liver Disease

Nonalcoholic fatty liver disease (NALFD) is now a leading cause of chronic liver disease worldwide, in part, as a consequence of rapidly rising levels of obesity and metabolic syndrome and is a major risk factor for cirrhosis, hepatocellular carcinoma, and liver-related mortality. From NAFLD stems a myriad of clinical challenges related to both diagnosis and management. A growing body of evidence suggests an intricate linkage between the gut microbiome and the pathogenesis of NAFLD. We highlight how our current knowledge of the gut-liver axis in NAFLD may be leveraged to develop gut microbiome-based personalized approaches for disease management, including its use as a non-invasive biomarker for diagnosis and staging, as a target for therapeutic modulation, and as a marker of drug response. We will also discuss current limitations of these microbiome-based approaches. Ultimately, a better understanding of microbiota-host interactions in NAFLD will inform the development of novel preventative strategies and precise therapeutic targets.

Maternal exposures and the infant gut microbiome: a systematic review with meta-analysis

Early life, including the establishment of the intestinal microbiome, represents a critical window of growth and development. Postnatal factors affecting the microbiome, including mode of delivery, feeding type, and antibiotic exposure have been widely investigated, but questions remain regarding the influence of exposures in utero on infant gut microbiome assembly. This systematic review aimed to synthesize evidence on exposures before birth, which affect the early intestinal microbiome. Five databases were searched in August 2019 for studies exploring pre-pregnancy or pregnancy 'exposure' data in relation to the infant microbiome. Of 1,441 publications identified, 76 were included. Factors reported influencing microbiome composition and diversity included maternal antibiotic and probiotic uses, dietary intake, pre-pregnancy body mass index (BMI), gestational weight gain (GWG), diabetes, mood, and others. Eleven studies contributed to three meta-analyses quantifying associations between maternal intrapartum antibiotic exposure (IAP), BMI and GWG, and infant microbiome alpha diversity (Shannon Index). IAP, maternal overweight/obesity and excessive GWG were all associated with reduced diversity. Most studies were observational, few included early recruitment or longitudinal follow-up, and the timing, frequency, and methodologies related to stool sampling and analysis were variable. Standardization and collaboration are imperative to enhance understanding in this complex and rapidly evolving area.

APP Medical Diagnostic Check-up Consultation System Based on Speech Recognition

Background:

Medical test orders can display the physiological functions of patients by using medical means. The medical staff determines the patient's condition through medical test orders and completes the treatment. However, for most patients and their families, there are so many terminologies in the medical test list and they are inconvenient to understand and query, which would affect the patients’ cognition and treatment effect. Therefore, it is especially necessary to develop a consulting system that can provide related analysis after getting medical test data.

Objective :

This paper starts with information acquisition and speech recognition. It proposes a natural scene information acquisition and analysis model based on deep learning, focusing on improving the recognition rate of routine test list and achieving targeted smart search to allow users to get more accurate personalized health advice.

Methods :

Based on medical characteristics, considering the needs of patients, this paper constructs an APP-based conventional medical test consultation system, using artificial intelligence and voice recognition technology to collect user input; analyzing user needs with the help of conventional medical information knowledge database.

Results:

This model combines speech recognition and data mining methods to obtain routine test list data and is suitable for accurate analysis of problems in routine check-up procedure. The app provides effective explanations and guidance for the treatment and rehabilitation of patients.

Conclusion:

It organically links the Internet with personalized medicine, which can effectively improve the popularity of medical knowledge and provide a reference for the application of medical services on the Internet. Meanwhile, this app can contribute to the improvement of medical standards and provide new models for modern medical management.

Impact of host genetics on gut microbiome: Take-home lessons from human and mouse studies

The intestinal microbiome has emerged as an important component involved in various diseases. Therefore, the interest in understanding the factors shaping its composition is growing. The gut microbiome, often defined as a complex trait, contains diverse components and its properties are determined by a combination of external and internal effects. Although much effort has been invested so far, it is still difficult to evaluate the extent to which human genetics shape the composition of the gut microbiota. However, in mouse studies, where the environmental factors are better controlled, the effect of the genetic background was significant. The purpose of this paper is to provide a current assessment of the role of human host genetics in shaping the gut microbiome composition. Despite the inconsistency of the reported results, it can be estimated that the genetic factor affects a portion of the microbiome. However, this effect is currently lower than the initial estimates, and it is difficult to separate the genetic influence from the environmental effect. Additionally, despite the differences between the microbial composition of humans and mice, results from mouse models can strengthen our knowledge of host genetics underlying the human gut microbial variation.

CoMNRank: An integrated approach to extract and prioritize human microbial metabolites from MEDLINE records

MOTIVATION

Trillions of bacteria in human body (human microbiota) affect human health and diseases by controlling host functions through small molecule metabolites.An accurate and comprehensive catalog of the metabolic output from human microbiota is critical for our deep understanding of how microbial metabolism contributes to human health.The large number of published biomedical research articles is a rich resource of microbiome studies.However, automatically extracting microbial metabolites from free-text documents and differentiating them from other human metabolites is a challenging task.Here we developed an integrated approach called Co-occurrence Metabolite Network Ranking (CoMNRank) by combining named entity extraction, network construction and topic sensitive network-based prioritization to extract and prioritize microbial metabolites from biomedical articles.

METHODS

The text data included 28,851,232 MEDLINE records.CoMNRank consists of three steps: (1) extraction of human metabolites from MEDLINE records; (2) construction of a weighted co-occurrence metabolite network (CoMN); (3) prioritization and differentiation of microbial metabolites from other human metabolites.

RESULTS

For the first step of CoMNRank, we extracted 11,846 human metabolites from MEDLINE articles, with a baseline performance of precision of 0.014, recall of 0.959 and F1 of 0.028.We then constructed a weighted CoMN of 6,996 nodes and 986,186 edges.CoMNRank effectively prioritized microbial metabolites: the precision of top ranked metabolites is 0.45, a 31-fold enrichment as compared to the overall precision of 0.014.Manual curation of top 100 metabolites showed a true precision of 0.67, among which 48% true positives are not captured by existing databases.

CONCLUSION

Our study sets the foundation for future tasks of microbial entity and relationship extractions as well as data-driven studies of how microbial metabolism contributes to human health and diseases.

A survey on predicting microbe-disease associations: biological data and computational methods

Various microbes have proved to be closely related to the pathogenesis of human diseases. While many computational methods for predicting human microbe-disease associations (MDAs) have been developed, few systematic reviews on these methods have been reported. In this study, we provide a comprehensive overview of the existing methods. Firstly, we introduce the data used in existing MDA prediction methods. Secondly, we classify those methods into different categories by their nature and describe their algorithms and strategies in detail. Next, experimental evaluations are conducted on representative methods using different similarity data and calculation methods to compare their prediction performances. Based on the principles of computational methods and experimental results, we discuss the advantages and disadvantages of those methods and propose suggestions for the improvement of prediction performances. Considering the problems of the MDA prediction at present stage, we discuss future work from three perspectives including data, methods and formulations at the end.

Knowledge-Based Biomedical Data Science

Knowledge-based biomedical data science involves the design and implementation of computer systems that act as if they knew about biomedicine. Such systems depend on formally represented knowledge in computer systems, often in the form of knowledge graphs. Here we survey recent progress in systems that use formally represented knowledge to address data science problems in both clinical and biological domains, as well as progress on approaches for creating knowledge graphs. Major themes include the relationships between knowledge graphs and machine learning, the use of natural language processing to construct knowledge graphs, and the expansion of novel knowledge-based approaches to clinical and biological domains.

Automatic extraction, prioritization and analysis of gut microbial metabolites from biomedical literature

Many diseases are driven by gene-environment interactions. One important environmental factor is the metabolic output of human gut microbiota. A comprehensive catalog of human metabolites originated in microbes is critical for data-driven approaches to understand how microbial metabolism contributes to human health and diseases. Here we present a novel integrated approach to automatically extract and analyze microbial metabolites from 28 million published biomedical records. First, we classified 28,851,232 MEDLINE records into microbial metabolism-related or not. Second, candidate microbial metabolites were extracted from the classified texts. Third, we developed signal prioritization algorithms to further differentiate microbial metabolites from metabolites originated from other resources. Finally, we systematically analyzed the interactions between extracted microbial metabolites and human genes. A total of 11,846 metabolites were extracted from 28 million MEDLINE articles. The combined text classification and signal prioritization significantly enriched true positives among top: manual curation of top 100 metabolites showed a true precision of 0.55, representing a significant 38.3-fold enrichment as compared to the precision of 0.014 for baseline extraction. More importantly, 29% extracted microbial metabolites have not been captured by existing databases. We performed data-driven analysis of the interactions between the extracted microbial metabolite and human genetics. This study represents the first effort towards automatically extracting and prioritizing microbial metabolites from published biomedical literature, which can set a foundation for future tasks of microbial metabolite relationship extraction from literature and facilitate data-driven studies of how microbial metabolism contributes to human diseases.

Gut microbiota and cardiovascular disease: opportunities and challenges

Coronary artery disease (CAD) is the most common health problem worldwide and remains the leading cause of morbidity and mortality. Over the past decade, it has become clear that the inhabitants of our gut, the gut microbiota, play a vital role in human metabolism, immunity, and reactions to diseases, including CAD. Although correlations have been shown between CAD and the gut microbiota, demonstration of potential causal relationships is much more complex and challenging. In this review, we will discuss the potential direct and indirect causal roots between gut microbiota and CAD development via microbial metabolites and interaction with the immune system. Uncovering the causal relationship of gut microbiota and CAD development can lead to novel microbiome-based preventative and therapeutic interventions. However, an interdisciplinary approach is required to shed light on gut bacterial-mediated mechanisms (e.g., using advanced nanomedicine technologies and incorporation of demographic factors such as age, sex, and ethnicity) to enable efficacious and high-precision preventative and therapeutic strategies for CAD.

OHMI: the ontology of host-microbiome interactions

BACKGROUND

Host-microbiome interactions (HMIs) are critical for the modulation of biological processes and are associated with several diseases. Extensive HMI studies have generated large amounts of data. We propose that the logical representation of the knowledge derived from these data and the standardized representation of experimental variables and processes can foster integration of data and reproducibility of experiments and thereby further HMI knowledge discovery.

METHODS

Through a multi-institutional collaboration, a community-based Ontology of Host-Microbiome Interactions (OHMI) was developed following the Open Biological/Biomedical Ontologies (OBO) Foundry principles. As an OBO library ontology, OHMI leverages established ontologies to create logically structured representations of (1) microbiomes, microbial taxonomy, host species, host anatomical entities, and HMIs under different conditions and (2) associated study protocols and types of data analysis and experimental results.

RESULTS

Aligned with the Basic Formal Ontology, OHMI comprises over 1000 terms, including terms imported from more than 10 existing ontologies together with some 500 OHMI-specific terms. A specific OHMI design pattern was generated to represent typical host-microbiome interaction studies. As one major OHMI use case, drawing on data from over 50 peer-reviewed publications, we identified over 100 bacteria and fungi from the gut, oral cavity, skin, and airway that are associated with six rheumatic diseases including rheumatoid arthritis. Our ontological study identified new high-level microbiota taxonomical structures. Two microbiome-related competency questions were also designed and addressed. We were also able to use OHMI to represent statistically significant results identified from a large existing microbiome database data analysis.

CONCLUSION

OHMI represents entities and relations in the domain of HMIs. It supports shared knowledge representation, data and metadata standardization and integration, and can be used in formulation of advanced queries for purposes of data analysis.