A nondegenerate code of deleterious variants in Mendelian loci contributes to complex disease risk.

Although countless highly penetrant variants have been associated with Mendelian disorders, the genetic etiologies underlying complex diseases remain largely unresolved. By mining the medical records of over 110 million patients, we examine the extent to which Mendelian variation contributes to complex disease risk. We detect thousands of associations between Mendelian and complex diseases, revealing a nondegenerate, phenotypic code that links each complex disorder to a unique collection of Mendelian loci. Using genome-wide association results, we demonstrate that common variants associated with complex diseases are enriched in the genes indicated by this "Mendelian code." Finally, we detect hundreds of comorbidity associations among Mendelian disorders, and we use probabilistic genetic modeling to demonstrate that Mendelian variants likely contribute nonadditively to the risk for a subset of complex diseases. Overall, this study illustrates a complementary approach for mapping complex disease loci and provides unique predictions concerning the etiologies of specific diseases.

Graph-based signal integration for high-throughput phenotyping.

BACKGROUND: Electronic Health Records aggregated in Clinical Data Warehouses (CDWs) promise to revolutionize Comparative Effectiveness Research and suggest new avenues of research. However, the effectiveness of CDWs is diminished by the lack of properly labeled data. We present a novel approach that integrates knowledge from the CDW, the biomedical literature, and the Unified Medical Language System (UMLS) to perform high-throughput phenotyping. In this paper, we automatically construct a graphical knowledge model and then use it to phenotype breast cancer patients. We compare the performance of this approach to using MetaMap when labeling records. RESULTS: MetaMap's overall accuracy at identifying breast cancer patients was 51.1% (n=428); recall=85.4%, precision=26.2%, and F1=40.1%. Our unsupervised graph-based high-throughput phenotyping had accuracy of 84.1%; recall=46.3%, precision=61.2%, and F1=52.8%. CONCLUSIONS: We conclude that our approach is a promising alternative for unsupervised high-throughput phenotyping.

Toward a veterinary informatics research agenda: an analysis of the PubMed-indexed literature.

PURPOSE: Veterinary medicine and human health are inextricably intertwined. Effective tracking of veterinary information - veterinary informatics - impacts not only veterinary medicine, but also public health, informatics research, and clinical care. However, veterinary informatics has received little attention from the general biomedical informatics community. METHODS: To identify both active and under-researched areas in veterinary informatics, we retrieved Medical Subject Heading (MeSH) descriptors for veterinary informatics-related citations and analyzed them by topic category, animal type, and journal. RESULTS: We found that the categories of veterinary informatics with the most growth were information/bibliographical retrieval, hardware/programming, and radiology/imaging. Less than two articles per year were published in the areas of computerized veterinary medical records, clinical decision support, standards, and controlled vocabularies. Veterinary informatics articles primarily address production animals such as cattle and sheep, and companion animals such as cats and dogs. Six journals account for 31% of the veterinary informatics literature, 35 journals account for 66%. CONCLUSIONS: Veterinary informatics remains an embryonic field with relatively few publications. With the exception of radiology/imaging, published articles are primarily focused on non-clinical areas such as hardware/programming and information retrieval. There are very few publications on controlled vocabularies, standards, methodologies for integrating disparate systems, computerized medical records, clinical decision support systems, and system usability. The lack of publications in these areas may hamper efforts to collect and track animal health data at a time when such data are potentially critical to human health.

Expert guided natural language processing using one-class classification.

INTRODUCTION: Automatically identifying specific phenotypes in free-text clinical notes is critically important for the reuse of clinical data. In this study, the authors combine expert-guided feature (text) selection with one-class classification for text processing. OBJECTIVES: To compare the performance of one-class classification to traditional binary classification; to evaluate the utility of feature selection based on expert-selected salient text (snippets); and to determine the robustness of these models with respects to irrelevant surrounding text. METHODS: The authors trained one-class support vector machines (1C-SVMs) and two-class SVMs (2C-SVMs) to identify notes discussing breast cancer. Manually annotated visit summary notes (88 positive and 88 negative for breast cancer) were used to compare the performance of models trained on whole notes labeled as positive or negative to models trained on expert-selected text sections (snippets) relevant to breast cancer status. Model performance was evaluated using a 70:30 split for 20 iterations and on a realistic dataset of 10 000 records with a breast cancer prevalence of 1.4%. RESULTS: When tested on a balanced experimental dataset, 1C-SVMs trained on snippets had comparable results to 2C-SVMs trained on whole notes (F = 0.92 for both approaches). When evaluated on a realistic imbalanced dataset, 1C-SVMs had a considerably superior performance (F = 0.61 vs. F = 0.17 for the best performing model) attributable mainly to improved precision (p = .88 vs. p = .09 for the best performing model). CONCLUSIONS: 1C-SVMs trained on expert-selected relevant text sections perform better than 2C-SVMs classifiers trained on either snippets or whole notes when applied to realistically imbalanced data with low prevalence of the positive class.

Duplicate patient records--implication for missed laboratory results.

INTRODUCTION: Although duplicate records are a potential patient safety hazard, the actual clinical harm associated with these records has never been studied. We hypothesized that duplicate records will be associated with missed abnormal laboratory results. METHODS: A retrospective, matched, cohort study of 904 events of abnormal laboratory result (HgbA1c, TSH, Vitamin B(12), LDL). We compared the rates of missed laboratory results between patients with duplicate and non-duplicate records from the ambulatory clinics. Cases were matched according to test and ordering physician. RESULTS: Duplicate records were associated with a higher rate of missed laboratory results (OR=1.44, 95% CI 1.1-1.9). Other factors associated with missed lab results were tests performed as screening (OR=2.22, 95% CI 1.4-3.4), and older age (OR=1.15 for every decade, 95% CI 1.01-1.2). In most cases test results were reported into the main patient record. DISCUSSION: Duplicate records were associated with a higher risk of missing important laboratory results.