Large-Scale Discovery of Disease-Disease and Disease-Gene Associations

Abstract

Data-driven phenotype analyses on Electronic Health Record (EHR) data have recently drawn benefits across many areas of clinical practice, uncovering new links in the medical sciences that can potentially affect the well-being of millions of patients. In this paper, EHR data is used to discover novel relationships between diseases by studying their comorbidities (co-occurrences in patients). A novel embedding model is designed to extract knowledge from disease comorbidities by learning from a large-scale EHR database comprising more than 35 million inpatient cases spanning nearly a decade, revealing significant improvements on disease phenotyping over current computational approaches. In addition, the use of the proposed methodology is extended to discover novel disease-gene associations by including valuable domain knowledge from genome-wide association studies. To evaluate our approach, its effectiveness is compared against a held-out set where, again, it revealed very compelling results. For selected diseases, we further identify candidate gene lists for which disease-gene associations were not studied previously. Thus, our approach provides biomedical researchers with new tools to filter genes of interest, thus, reducing costly lab studies.

Human disease cost network analysis

“…Various data sources have been used in multidisease research. Some studies use hospital‐ or community‐based data, and may have a sample selection bias problem . Some others are based on large insurance systems and use claim data .…”

“…Some studies use hospital-or community-based data, and may have a sample selection bias problem. 6,11,15 Some others are based on large insurance systems and use claim data. 10 However, most insurance systems, such as Medicare and Kaiser, also have a sample selection bias problem and cannot describe disease properties for the general population.…”

Omics‐based Identification of Glycan Structures as Biomarkers for a Variety of Diseases

“…Some proteins and lipids are terminally expressed as glycoproteins and glycolipids, which determines their final biological activity. [5,6,7] In recent years, genomic data analysis has become a popular approach to deepen our understanding of the molecular mechanisms of a disease of interest. Glycans can serve as intermediates in energy generation, as signaling molecules, or as structural components, and glycan structures are key factors in interactions with proteins (e. g., glycoprotein-lectin), cell-cell communication, and host-pathogen interactions.…”

Integrated molecular, clinical, and ontological analysis identifies overlooked disease relationships

“…1, 2 However, existing knowledge of disease relationships is incomplete and flawed. Previous research exploring molecular disease relationships have used genetic (genome wide associations studies 3, 4, 5, 6, 7, 8,9,10 Figure 1: Integrated molecular, clinical, and ontological analysis for insight into disease similarity. Transcriptomic meta-analysis was performed to calculate effect sizes, a measure of differential expression, of genes across 104 diseases or conditions.…”

“…Analyses that examine relationships among human diseases at multiple different scales have the potential to significantly improve our ability to understand disease etiology, postulate therapeutic targets, and connect disparate research communities (1,2). Molecular relationships among diseases have primarily been studied in terms of the genetic similarities derived from either genome wide association studies (GWAS) (3)(4)(5)(6)(7)(8)(9)(10) or, more recently, phenome wide association studies (PheWAS) (11)(12)(13). By identifying patterns of genetic architecture that are shared by diseases, such studies can answer questions about the genetic origin of diseases and explain the seemingly unrelated phenotypic effects of genetic variation (14).…”