Uncovering exposures responsible for birth season - disease effects: a global study

Objective

Birth month and climate impact lifetime disease risk, while the underlying exposures remain largely elusive. We seek to uncover distal risk factors underlying these relationships by probing the relationship between global exposure variance and disease risk variance by birth season.

Material and Methods

This study utilizes electronic health record data from 6 sites representing 10.5 million individuals in 3 countries (United States, South Korea, and Taiwan). We obtained birth month–disease risk curves from each site in a case-control manner. Next, we correlated each birth month–disease risk curve with each exposure. A meta-analysis was then performed of correlations across sites. This allowed us to identify the most significant birth month–exposure relationships supported by all 6 sites while adjusting for multiplicity. We also successfully distinguish relative age effects (a cultural effect) from environmental exposures.

Results

Attention deficit hyperactivity disorder was the only identified relative age association. Our methods identified several culprit exposures that correspond well with the literature in the field. These include a link between first-trimester exposure to carbon monoxide and increased risk of depressive disorder (R = 0.725, confidence interval [95% CI], 0.529-0.847), first-trimester exposure to fine air particulates and increased risk of atrial fibrillation (R = 0.564, 95% CI, 0.363-0.715), and decreased exposure to sunlight during the third trimester and increased risk of type 2 diabetes mellitus (R = −0.816, 95% CI, −0.5767, −0.929).

Conclusion

A global study of birth month–disease relationships reveals distal risk factors involved in causal biological pathways that underlie them.

Climate Classification is an Important Factor in Assessing Quality-of-Care Across Hospitals

Climate is a known modulator of disease, but its impact on hospital performance metrics remains unstudied. We assess the relationship between Köppen-Geiger climate classification and hospital performance metrics, specifically 30-day mortality, as reported in Hospital Compare, and collected for the period July 2013 through June 2014 (7/1/2013-06/30/2014). A hospital-level multivariate linear regression analysis was performed while controlling for known socioeconomic factors to explore the relationship between all-cause mortality and climate. Hospital performance scores were obtained from 4,524 hospitals belonging to 15 distinct Köppen-Geiger climates and 2,373 unique counties. Model results revealed that hospital performance metrics for mortality showed significant climate dependence (p < 0.001) after adjusting for socioeconomic factors. Climate is a significant factor in evaluating hospital 30-day mortality rates. These results demonstrate that climate classification is an important factor when comparing hospital performance across the United States.

Coupling Data Mining and Laboratory Experiments to Discover Drug Interactions Causing QT Prolongation

BACKGROUND

QT interval-prolonging drug-drug interactions (QT-DDIs) may increase the risk of life-threatening arrhythmia. Despite guidelines for testing from regulatory agencies, these interactions are usually discovered after drugs are marketed and may go undiscovered for years.

OBJECTIVES

Using a combination of adverse event reports, electronic health records (EHR), and laboratory experiments, the goal of this study was to develop a data-driven pipeline for discovering QT-DDIs.

METHODS

1.8 million adverse event reports were mined for signals indicating a QT-DDI. Using 1.6 million electrocardiogram results from 380,000 patients in our institutional EHR, these putative interactions were either refuted or corroborated. In the laboratory, we used patch-clamp electrophysiology to measure the human ether-à-go-go-related gene (hERG) channel block (the primary mechanism by which drugs prolong the QT interval) to evaluate our top candidate.

RESULTS

Both direct and indirect signals in the adverse event reports provided evidence that the combination of ceftriaxone (a cephalosporin antibiotic) and lansoprazole (a proton-pump inhibitor) will prolong the QT interval. In the EHR, we found that patients taking both ceftriaxone and lansoprazole had significantly longer QTc intervals (up to 12 ms in white men) and were 1.4 times more likely to have a QTc interval above 500 ms. In the laboratory, we found that, in combination and at clinically relevant concentrations, these drugs blocked the hERG channel. As a negative control, we evaluated the combination of lansoprazole and cefuroxime (another cephalosporin), which lacked evidence of an interaction in the adverse event reports. We found no significant effect of this pair in either the EHR or in the electrophysiology experiments. Class effect analyses suggested this interaction was specific to lansoprazole combined with ceftriaxone but not with other cephalosporins.

CONCLUSIONS

Coupling data mining and laboratory experiments is an efficient method for identifying QT-DDIs. Combination therapy of ceftriaxone and lansoprazole is associated with increased risk of acquired long QT syndrome.

PREDICTIVE MODELING OF HOSPITAL READMISSION RATES USING ELECTRONIC MEDICAL RECORD-WIDE MACHINE LEARNING: A CASE-STUDY USING MOUNT SINAI HEART FAILURE COHORT

Reduction of preventable hospital readmissions that result from chronic or acute conditions like stroke, heart failure, myocardial infarction and pneumonia remains a significant challenge for improving the outcomes and decreasing the cost of healthcare delivery in the United States. Patient readmission rates are relatively high for conditions like heart failure (HF) despite the implementation of high-quality healthcare delivery operation guidelines created by regulatory authorities. Multiple predictive models are currently available to evaluate potential 30-day readmission rates of patients. Most of these models are hypothesis driven and repetitively assess the predictive abilities of the same set of biomarkers as predictive features. In this manuscript, we discuss our attempt to develop a data-driven, electronic-medical record-wide (EMR-wide) feature selection approach and subsequent machine learning to predict readmission probabilities. We have assessed a large repertoire of variables from electronic medical records of heart failure patients in a single center. The cohort included 1,068 patients with 178 patients were readmitted within a 30-day interval (16.66% readmission rate). A total of 4,205 variables were extracted from EMR including diagnosis codes (n=1,763), medications (n=1,028), laboratory measurements (n=846), surgical procedures (n=564) and vital signs (n=4). We designed a multistep modeling strategy using the Naïve Bayes algorithm. In the first step, we created individual models to classify the cases (readmitted) and controls (non-readmitted). In the second step, features contributing to predictive risk from independent models were combined into a composite model using a correlation-based feature selection (CFS) method. All models were trained and tested using a 5-fold cross-validation method, with 70% of the cohort used for training and the remaining 30% for testing. Compared to existing predictive models for HF readmission rates (AUCs in the range of 0.6-0.7), results from our EMR-wide predictive model (AUC=0.78; Accuracy=83.19%) and phenome-wide feature selection strategies are encouraging and reveal the utility of such datadriven machine learning. Fine tuning of the model, replication using multi-center cohorts and prospective clinical trial to evaluate the clinical utility would help the adoption of the model as a clinical decision system for evaluating readmission status.

METHODS TO ENSURE THE REPRODUCIBILITY OF BIOMEDICAL RESEARCH

Science is not done in a vacuum - across fields of biomedicine, scientists have built on previous research and used data published in previous papers. A mainstay of scientific inquiry is the publication of one's research and recognition for this work is given in the form of citations and notoriety - ideally given in proportion to the quality of the work. Academic incentives, however, may encourage individual researchers to prioritize career ambitions over scientific truth. Recently, the New England Journal of Medicine published a commentary calling scientists who repurpose data "research parasites" who misuse data generated by others to demonstrate alternative hypotheses. In our opinion, the concept of data hoarding not only runs contrary to the spirit of, but also hinders scientific progress. Scientific research is meant to seek objective truth, rather than promote a personal agenda, and the only way to do so is through maximum transparency and reproducibility, no matter who is using the data….

An Integrative Data Science Pipeline to Identify Novel Drug Interactions that Prolong the QT Interval

Introduction

Drug-induced prolongation of the QT interval on the electrocardiogram (long QT syndrome, LQTS) can lead to a potentially fatal ventricular arrhythmia known as torsades de pointes (TdP). Over 40 drugs with both cardiac and non-cardiac indications are associated with increased risk of TdP, but drug–drug interactions contributing to LQTS (QT-DDIs) remain poorly characterized. Traditional methods for mining observational healthcare data are poorly equipped to detect QT-DDI signals due to low reporting numbers and lack of direct evidence for LQTS.

Objective

We hypothesized that LQTS could be identified latently using an adverse event (AE) fingerprint of more commonly reported AEs. We aimed to generate an integrated data science pipeline that addresses current limitations by identifying latent signals for QT-DDIs in the US FDA’s Adverse Event Reporting System (FAERS) and retrospectively validating these predictions using electrocardiogram data in electronic health records (EHRs).

Methods

We trained a model to identify an AE fingerprint for risk of TdP for single drugs and applied this model to drug pair data to predict novel DDIs. In the EHR at Columbia University Medical Center, we compared the QTc intervals of patients prescribed the flagged drug pairs with patients prescribed either drug individually.

Results

We created an AE fingerprint consisting of 13 latently detected side effects. This model significantly outperformed a direct evidence control model in the detection of established interactions (p = 1.62E−3) and significantly enriched for validated QT-DDIs in the EHR (p = 0.01). Of 889 pairs flagged in FAERS, eight novel QT-DDIs were significantly associated with prolonged QTc intervals in the EHR and were not due to co-prescribed medications.

Conclusions

Latent signal detection in FAERS validated using the EHR presents an automated and data-driven approach for systematically identifying novel QT-DDIs. The high-confidence hypotheses flagged using this method warrant further investigation.

Electronic supplementary material

The online version of this article (doi:10.1007/s40264-016-0393-1) contains supplementary material, which is available to authorized users.

A Drug Safety Rating System Based on Postmarketing Costs Associated with Adverse Events and Patient Outcomes

BACKGROUND

Given the multiple limitations associated with relatively homogeneous preapproval clinical trials, inadequate data disclosures, slow reaction times from regulatory bodies, and deep-rooted bias against disclosing and publishing negative results, there is an acute need for the development of analytics that reflect drug safety in heterogeneous, real-world populations.

OBJECTIVE

To develop a drug safety statistic that estimates downstream medical costs associated with serious adverse events (AEs) and unfavorable patient outcomes associated with the use of 706 FDA-approved drugs.

METHODS

All primary suspect case reports for each drug were collected from the FDA's Adverse Event Reporting System database (FAERS) from 2010-2014. The Medical Dictionary for Regulatory Activities (MedDRA) was used to code serious AEs and outcomes, which were tallied for each case report. Medical costs associated with AEs and poor patient outcomes were derived from Agency for Healthcare Research and Quality (AHRQ) survey data, and their corresponding ICD-9-CM codes were mapped to MedDRA terms. Nonserious AEs and outcomes were not included. For each case report, either the highest AE cost or, if no eligible AE was listed, the highest outcome cost was used. All costed cases were aggregated for each drug and divided by the number of patients exposed to obtain a downstream estimated direct medical cost burden per exposure. Each drug was assigned a corresponding 1-100 point total.

RESULTS

The 706 drugs showed an exponential distribution of downstream costs, and the data were transformed using the natural log to approximate a normal distribution. The minimum score was 8.29, and the maximum score was 99.25, with a mean of 44.32. Drugs with the highest individual scores tended to be kinase inhibitors, thalidomide analogs, and endothelin receptor antagonists. When scores were analyzed across Established Pharmacologic Class (EPC), the kinase inhibitor and endothelin receptor antagonist classes had the highest total. However, other EPCs with median scores of 75 and above included hepatitis C virus NS3/4A protease inhibitor, recombinant human interferon beta, vascular endothelial growth factor-directed antibody, and tumor necrosis factor blocker. When Anatomical Therapeutic Chemical classifications were analyzed, antineoplastic drugs were outliers with approximately 80% of their individual scores 60 and above, while approximately 20%-30% of blood and anti-infective drugs had scores of 60 and above. Within-drug class results served to differentiate similar drugs. For example, 6 serotonin reuptake inhibitors had a score range of 35 to 53.

CONCLUSIONS

This scoring system is based on estimated direct medical costs associated with postmarketing AEs and poor patient outcomes and thereby helps fill a large information gap regarding drug safety in real-world patient populations.

Predicting G protein-coupled receptor downstream signaling by tissue expression

MOTIVATION

G protein-coupled receptors (GPCRs) are central to how cells respond to their environment and a major class of pharmacological targets. However, comprehensive knowledge of which pathways are activated and deactivated by these essential sensors is largely unknown. To better understand the mechanism of GPCR signaling system, we integrated five independent genome-wide expression datasets, representing 275 human tissues and cell lines, with protein-protein interactions and functional pathway data.

RESULTS

We found that tissue-specificity plays a crucial part in the function of GPCR signaling system. Only a few GPCRs are expressed in each tissue, which are coupled by different combinations of G-proteins or β-arrestins to trigger specific downstream pathways. Based on this finding, we predicted the downstream pathways of GPCR in human tissues and validated our results with L1000 knockdown data. In total, we identified 154,988 connections between 294 GPCRs and 690 pathways in 240 tissues and cell types.

AVAILABILITY AND IMPLEMENTATION

The source code and results supporting the conclusions of this article are available at http://tatonettilab.org/resources/GOTE/source_code/ CONTACT: nick.tatonetti@columbia.eduSupplementary information: Supplementary data are available at Bioinformatics online.

Using a Novel Ontology to Inform the Discovery of Therapeutic Peptides from Animal Venoms

Venoms and venom-derived compounds constitute a rich and largely unexplored source of potentially therapeutic compounds. To facilitate biomedical research, it is necessary to design a robust informatics infrastructure that will allow semantic computation of venom concepts in a standardized, consistent manner. We have designed an ontology of venom-related concepts - named Venom Ontology - that reuses an existing public data source: UniProt's Tox-Prot database. In addition to describing the ontology and its construction, we have performed three separate case studies demonstrating its utility: (1) An exploration of venom peptide similarity networks within specific genera; (2) A broad overview of the distribution of available data among common taxonomic groups spanning the known tree of life; and (3) An analysis of the distribution of venom complexity across those same taxonomic groups. Venom Ontology is publicly available on BioPortal at http://bioportal.bioontology.org/ontologies/CU-VO.

Investigation of 7-dehydrocholesterol reductase pathway to elucidate off-target prenatal effects of pharmaceuticals: a systematic review

Mendelian diseases contain important biological information regarding developmental effects of gene mutations that can guide drug discovery and toxicity efforts. In this review, we focus on Smith–Lemli–Opitz syndrome (SLOS), a rare Mendelian disease characterized by compound heterozygous mutations in 7-dehydrocholesterol reductase (DHCR7) resulting in severe fetal deformities. We present a compilation of SLOS-inducing DHCR7 mutations and the geographic distribution of those mutations in healthy and diseased populations. We observed that several mutations thought to be disease causing occur in healthy populations, indicating an incomplete understanding of the condition and highlighting new research opportunities. We describe the functional environment around DHCR7, including pharmacological DHCR7 inhibitors and cholesterol and vitamin D synthesis. Using PubMed, we investigated the fetal outcomes following prenatal exposure to DHCR7 modulators. First-trimester exposure to DHCR7 inhibitors resulted in outcomes similar to those of known teratogens (50 vs 48% born-healthy). DHCR7 activity should be considered during drug development and prenatal toxicity assessment.

A Computational Drug Repositioning Approach for Targeting Oncogenic Transcription Factors

Mutations in transcription factor (TF) genes are frequently observed in tumors, often leading to aberrant transcriptional activity. Unfortunately, TFs are often considered undruggable due to the absence of targetable enzymatic activity. To address this problem, we developed CRAFTT, a computational drug-repositioning approach for targeting TF activity. CRAFTT combines ChIP-seq with drug-induced expression profiling to identify small molecules that can specifically perturb TF activity. Application to ENCODE ChIP-seq datasets revealed known drug-TF interactions, and a global drug-protein network analysis supported these predictions. Application of CRAFTT to ERG, a pro-invasive, frequently overexpressed oncogenic TF, predicted that dexamethasone would inhibit ERG activity. Dexamethasone significantly decreased cell invasion and migration in an ERG-dependent manner. Furthermore, analysis of electronic medical record data indicates a protective role for dexamethasone against prostate cancer. Altogether, our method provides a broadly applicable strategy for identifying drugs that specifically modulate TF activity.

A curated and standardized adverse drug event resource to accelerate drug safety research

Identification of adverse drug reactions (ADRs) during the post-marketing phase is one of the most important goals of drug safety surveillance. Spontaneous reporting systems (SRS) data, which are the mainstay of traditional drug safety surveillance, are used for hypothesis generation and to validate the newer approaches. The publicly available US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) data requires substantial curation before they can be used appropriately, and applying different strategies for data cleaning and normalization can have material impact on analysis results. We provide a curated and standardized version of FAERS removing duplicate case records, applying standardized vocabularies with drug names mapped to RxNorm concepts and outcomes mapped to SNOMED-CT concepts, and pre-computed summary statistics about drug-outcome relationships for general consumption. This publicly available resource, along with the source code, will accelerate drug safety research by reducing the amount of time spent performing data management on the source FAERS reports, improving the quality of the underlying data, and enabling standardized analyses using common vocabularies.

Improving condition severity classification with an efficient active learning based framework

Classification of condition severity can be useful for discriminating among sets of conditions or phenotypes, for example when prioritizing patient care or for other healthcare purposes. Electronic Health Records (EHRs) represent a rich source of labeled information that can be harnessed for severity classification. The labeling of EHRs is expensive and in many cases requires employing professionals with high level of expertise. In this study, we demonstrate the use of Active Learning (AL) techniques to decrease expert labeling efforts. We employ three AL methods and demonstrate their ability to reduce labeling efforts while effectively discriminating condition severity. We incorporate three AL methods into a new framework based on the original CAESAR (Classification Approach for Extracting Severity Automatically from Electronic Health Records) framework to create the Active Learning Enhancement framework (CAESAR-ALE). We applied CAESAR-ALE to a dataset containing 516 conditions of varying severity levels that were manually labeled by seven experts. Our dataset, called the "CAESAR dataset," was created from the medical records of 1.9 million patients treated at Columbia University Medical Center (CUMC). All three AL methods decreased labelers' efforts compared to the learning methods applied by the original CAESER framework in which the classifier was trained on the entire set of conditions; depending on the AL strategy used in the current study, the reduction ranged from 48% to 64% that can result in significant savings, both in time and money. As for the PPV (precision) measure, CAESAR-ALE achieved more than 13% absolute improvement in the predictive capabilities of the framework when classifying conditions as severe. These results demonstrate the potential of AL methods to decrease the labeling efforts of medical experts, while increasing accuracy given the same (or even a smaller) number of acquired conditions. We also demonstrated that the methods included in the CAESAR-ALE framework (Exploitation and Combination_XA) are more robust to the use of human labelers with different levels of professional expertise.

Mapping the effects of drugs on the immune system

Understanding how drugs affect the immune system has consequences for treating disease and minimizing unwanted side effects. Here we present an integrative computational approach for predicting interactions between drugs and immune cells in a system-wide manner. The approach matches gene sets between transcriptional signatures to determine their similarity. We apply the method to model the interactions between 1,309 drugs and 221 immune cell types and predict 69,995 known and novel interactions. The resulting immune-cell pharmacology map is used to predict how 5 drugs influence 4 immune cell types in humans and mice. To validate the predictions, we analyzed patient records and examined cell population changes from in vivo experiments. Our method offers a tool for screening thousands of interactions to identify relationships between drugs and the immune system.

VenomKB, a new knowledge base for facilitating the validation of putative venom therapies

Animal venoms have been used for therapeutic purposes since the dawn of recorded history. Only a small fraction, however, have been tested for pharmaceutical utility. Modern computational methods enable the systematic exploration of novel therapeutic uses for venom compounds. Unfortunately, there is currently no comprehensive resource describing the clinical effects of venoms to support this computational analysis. We present VenomKB, a new publicly accessible knowledge base and website that aims to act as a repository for emerging and putative venom therapies. Presently, it consists of three database tables: (1) Manually curated records of putative venom therapies supported by scientific literature, (2) automatically parsed MEDLINE articles describing compounds that may be venom derived, and their effects on the human body, and (3) automatically retrieved records from the new Semantic Medline resource that describe the effects of venom compounds on mammalian anatomy. Data from VenomKB may be selectively retrieved in a variety of popular data formats, are open-source, and will be continually updated as venom therapies become better understood.

Systems biology approaches for identifying adverse drug reactions and elucidating their underlying biological mechanisms

Small molecules are indispensable to modern medical therapy. However, their use may lead to unintended, negative medical outcomes commonly referred to as adverse drug reactions (ADRs). These effects vary widely in mechanism, severity, and populations affected, making ADR prediction and identification important public health concerns. Current methods rely on clinical trials and postmarket surveillance programs to find novel ADRs; however, clinical trials are limited by small sample size, whereas postmarket surveillance methods may be biased and inherently leave patients at risk until sufficient clinical evidence has been gathered. Systems pharmacology, an emerging interdisciplinary field combining network and chemical biology, provides important tools to uncover and understand ADRs and may mitigate the drawbacks of traditional methods. In particular, network analysis allows researchers to integrate heterogeneous data sources and quantify the interactions between biological and chemical entities. Recent work in this area has combined chemical, biological, and large-scale observational health data to predict ADRs in both individual patients and global populations. In this review, we explore the rapid expansion of systems pharmacology in the study of ADRs. We enumerate the existing methods and strategies and illustrate progress in the field with a model framework that incorporates crucial data elements, such as diet and comorbidities, known to modulate ADR risk. Using this framework, we highlight avenues of research that may currently be underexplored, representing opportunities for future work.

An Assessment of Family History Information Captured in an Electronic Health Record

Family history is considered a core element of clinical care. In this study we assessed the quality of family history data captured in an established commercial electronic health record (EHR) at a large academic medical center. Because the EHR had no centralized location to store family history information, it was collected as part of clinical notes in structured or free-text format. We analyzed differences between 10,000 free-text and 9,121 structured family history observations. Each observation was classified according to disease presence/absence and family member affected (e.g., father, mother, etc.). The structured notes did not collect a complete family history as defined by standards endorsed by the U.S. Agency for Healthcare Research and Quality; the free-text notes contained more information than the structured notes, but still not enough to be considered "complete." Several barriers remain for collecting complete, useful family history data in electronic health records.

Connectivity Homology Enables Inter-Species Network Models of Synthetic Lethality

Synthetic lethality is a genetic interaction wherein two otherwise nonessential genes cause cellular inviability when knocked out simultaneously. Drugs can mimic genetic knock-out effects; therefore, our understanding of promiscuous drugs, polypharmacology-related adverse drug reactions, and multi-drug therapies, especially cancer combination therapy, may be informed by a deeper understanding of synthetic lethality. However, the colossal experimental burden in humans necessitates in silico methods to guide the identification of synthetic lethal pairs. Here, we present SINaTRA (Species-INdependent TRAnslation), a network-based methodology that discovers genome-wide synthetic lethality in translation between species. SINaTRA uses connectivity homology, defined as biological connectivity patterns that persist across species, to identify synthetic lethal pairs. Importantly, our approach does not rely on genetic homology or structural and functional similarity, and it significantly outperforms models utilizing these data. We validate SINaTRA by predicting synthetic lethality in S. pombe using S. cerevisiae data, then identify over one million putative human synthetic lethal pairs to guide experimental approaches. We highlight the translational applications of our algorithm for drug discovery by identifying clusters of genes significantly enriched for single- and multi-drug cancer therapies.

Synthetic lethality is a genetic interaction that has promising implications for informing novel cancer therapies. Over 200 million pairwise tests would be required to identify all pairwise synthetic lethal interactions in humans–currently, an impossibly large experimental burden. To simplify the process, we have developed a method to predict human synthetic lethal pairs in translation from a well-studied species to one in which synthetic lethality is understudied using both species’ protein-protein interaction networks. Here, we explore the model’s success in translation from S. cerevisiae to S. pombe. We then predict human synthetic lethality and suggest novel areas of inquiry for cancer therapies.

Improving Detection of Arrhythmia Drug-Drug Interactions in Pharmacovigilance Data through the Implementation of Similarity-Based Modeling

Identification of Drug-Drug Interactions (DDIs) is a significant challenge during drug development and clinical practice. DDIs are responsible for many adverse drug effects (ADEs), decreasing patient quality of life and causing higher care expenses. DDIs are not systematically evaluated in pre-clinical or clinical trials and so the FDA U. S. Food and Drug Administration relies on post-marketing surveillance to monitor patient safety. However, existing pharmacovigilance algorithms show poor performance for detecting DDIs exhibiting prohibitively high false positive rates. Alternatively, methods based on chemical structure and pharmacological similarity have shown promise in adverse drug event detection. We hypothesize that the use of chemical biology data in a post hoc analysis of pharmacovigilance results will significantly improve the detection of dangerous interactions. Our model integrates a reference standard of DDIs known to cause arrhythmias with drug similarity data. To compare similarity between drugs we used chemical structure (both 2D and 3D molecular structure), adverse drug side effects, chemogenomic targets, drug indication classes, and known drug-drug interactions. We evaluated the method on external reference standards. Our results showed an enhancement of sensitivity, specificity and precision in different top positions with the use of similarity measures to rank the candidates extracted from pharmacovigilance data. For the top 100 DDI candidates, similarity-based modeling yielded close to twofold precision enhancement compared to the proportional reporting ratio (PRR). Moreover, the method helps in the DDI decision making through the identification of the DDI in the reference standard that generated the candidate.

Birth month affects lifetime disease risk: a phenome-wide method

OBJECTIVE

An individual's birth month has a significant impact on the diseases they develop during their lifetime. Previous studies reveal relationships between birth month and several diseases including atherothrombosis, asthma, attention deficit hyperactivity disorder, and myopia, leaving most diseases completely unexplored. This retrospective population study systematically explores the relationship between seasonal affects at birth and lifetime disease risk for 1688 conditions.

METHODS

We developed a hypothesis-free method that minimizes publication and disease selection biases by systematically investigating disease-birth month patterns across all conditions. Our dataset includes 1 749 400 individuals with records at New York-Presbyterian/Columbia University Medical Center born between 1900 and 2000 inclusive. We modeled associations between birth month and 1688 diseases using logistic regression. Significance was tested using a chi-squared test with multiplicity correction.

RESULTS

We found 55 diseases that were significantly dependent on birth month. Of these 19 were previously reported in the literature (P < .001), 20 were for conditions with close relationships to those reported, and 16 were previously unreported. We found distinct incidence patterns across disease categories.

CONCLUSIONS

Lifetime disease risk is affected by birth month. Seasonally dependent early developmental mechanisms may play a role in increasing lifetime risk of disease.

Toward a complete dataset of drug-drug interaction information from publicly available sources

Although potential drug-drug interactions (PDDIs) are a significant source of preventable drug-related harm, there is currently no single complete source of PDDI information. In the current study, all publically available sources of PDDI information that could be identified using a comprehensive and broad search were combined into a single dataset. The combined dataset merged fourteen different sources including 5 clinically-oriented information sources, 4 Natural Language Processing (NLP) Corpora, and 5 Bioinformatics/Pharmacovigilance information sources. As a comprehensive PDDI source, the merged dataset might benefit the pharmacovigilance text mining community by making it possible to compare the representativeness of NLP corpora for PDDI text extraction tasks, and specifying elements that can be useful for future PDDI extraction purposes. An analysis of the overlap between and across the data sources showed that there was little overlap. Even comprehensive PDDI lists such as DrugBank, KEGG, and the NDF-RT had less than 50% overlap with each other. Moreover, all of the comprehensive lists had incomplete coverage of two data sources that focus on PDDIs of interest in most clinical settings. Based on this information, we think that systems that provide access to the comprehensive lists, such as APIs into RxNorm, should be careful to inform users that the lists may be incomplete with respect to PDDIs that drug experts suggest clinicians be aware of. In spite of the low degree of overlap, several dozen cases were identified where PDDI information provided in drug product labeling might be augmented by the merged dataset. Moreover, the combined dataset was also shown to improve the performance of an existing PDDI NLP pipeline and a recently published PDDI pharmacovigilance protocol. Future work will focus on improvement of the methods for mapping between PDDI information sources, identifying methods to improve the use of the merged dataset in PDDI NLP algorithms, integrating high-quality PDDI information from the merged dataset into Wikidata, and making the combined dataset accessible as Semantic Web Linked Data.

Development and validation of a classification approach for extracting severity automatically from electronic health records

Background

Electronic Health Records (EHRs) contain a wealth of information useful for studying clinical phenotype-genotype relationships. Severity is important for distinguishing among phenotypes; however other severity indices classify patient-level severity (e.g., mild vs. acute dermatitis) rather than phenotype-level severity (e.g., acne vs. myocardial infarction). Phenotype-level severity is independent of the individual patient’s state and is relative to other phenotypes. Further, phenotype-level severity does not change based on the individual patient. For example, acne is mild at the phenotype-level and relative to other phenotypes. Therefore, a given patient may have a severe form of acne (this is the patient-level severity), but this does not effect its overall designation as a mild phenotype at the phenotype-level.

Methods

We present a method for classifying severity at the phenotype-level that uses the Systemized Nomenclature of Medicine – Clinical Terms. Our method is called the Classification Approach for Extracting Severity Automatically from Electronic Health Records (CAESAR). CAESAR combines multiple severity measures – number of comorbidities, medications, procedures, cost, treatment time, and a proportional index term. CAESAR employs a random forest algorithm and these severity measures to discriminate between severe and mild phenotypes.

Results

Using a random forest algorithm and these severity measures as input, CAESAR differentiates between severe and mild phenotypes (sensitivity = 91.67, specificity = 77.78) when compared to a manually evaluated reference standard (k = 0.716).

Conclusions

CAESAR enables researchers to measure phenotype severity from EHRs to identify phenotypes that are important for comparative effectiveness research.

Are All Vaccines Created Equal? Using Electronic Health Records to Discover Vaccines Associated With Clinician-Coded Adverse Events

Adverse drug events (ADEs) are responsible for unnecessary patient deaths making them a major public health issue. Literature estimates 1% of ADEs recorded in Electronic Health Records (EHRs) are reported to federal databases making EHRs a vital source of ADE-related information. Using Columbia University Medical Center (CUMC)'s EHRs, we developed an algorithm to mine for vaccine-related ADEs occurring within 3 months of vaccination. In phase one, we measured the association between vaccinated patients with an ADE (cases) against those vaccinated without an ADE. To adjust for healthcare-process effects, phase two compared cases against those who returned to CUMC within 3 months without an ADE. We report 7 results passing multiplicity correction after demographic confounder adjustment. We observed an association, having some literature support, between swine flu vaccination and ADEs (H1N1v-like, OR=9.469, p<0.001; H1N1/H3N2, OR=3.207, p<0.001). Our algorithm could inform clinicians of the risks/benefits of vaccinations towards improving clinical care.

A knowledge-based, automated method for phenotyping in the EHR using only clinical pathology reports

The secondary use of electronic health records (EHR) represents unprecedented opportunities for biomedical discovery. Central to this goal is, EHR-phenotyping, also known as cohort identification, which remains a significant challenge. Complex phenotypes often require multivariate and multi-scale analyses, ultimately leading to manually created phenotype definitions. We present Ontology-driven Reports-based Phenotyping from Unique Signatures (ORPheUS), an automated approach to EHR-phenotyping. To do this we identify unique signatures of abnormal clinical pathology reports that correspond to pre-defined medical terms from biomedical ontologies. By using only the clinical pathology, or "lab", reports we are able to mitigate clinical biases enabling researchers to explore other dimensions of the EHR. We used ORPheUS to generate signatures for 858 diseases and validated against reference cohorts for Type 2 Diabetes Mellitus (T2DM) and Atrial Fibrillation (AF). Our results suggest that our approach, using solely clinical pathology reports, is an effective as a primary screening tool for automated clinical phenotyping.

The Weber effect and the United States Food and Drug Administration's Adverse Event Reporting System (FAERS): analysis of sixty-two drugs approved from 2006 to 2010

Background

The United States Food and Drug Administration’s (FDA) Adverse Event Reporting System (FAERS) consists of adverse event (AE) reports linked to approved drugs. The database is widely used to support post-marketing safety surveillance programs. Sometimes cited as a limitation to the usefulness of FAERS, however, is the ‘Weber effect,’ which is often summarized by stating that AE reporting peaks at the end of the second year after a regulatory authority approves a drug. Weber described this effect in 1984 based upon a single class of medications prescribed in the United Kingdom. Since that time, the FDA has made a concerted effort to improve both reporting and the database itself. Both volume and quality of AE reporting has dramatically improved since Weber’s report, with an estimated 800,000 yearly reports now being logged into FAERS.

Objective

The aim of this study was to determine if current FAERS reporting follows the trend described by Weber.

Methods

Sixty-two drugs approved by the FDA between 2006 and 2010 were included in this analysis. Publicly available FAERS data were used to assess the ‘primary suspect’ AE reporting pattern for up to a 4-year period following each drug’s approval date.

Results

A total of 334,984 AE reports were logged into FAERS for the 62 drugs analyzed here. While a few of the drugs demonstrated what could be considered ‘Weber effect’ curves, a majority of the drugs showed little evidence for the effect. In fact, the general AE reporting pattern observed in this study appears to consist simply of increasing case counts over the first three quarters after approval followed by relatively constant counts thereafter.

Conclusions

Our results suggest that most of the modern adverse event reporting into FAERS does not follow the pattern described by Weber. Factors that may have contributed to this finding include large increases in the volume of AE reports since the Weber effect was described, as well as a concerted effort by the FDA to increase awareness regarding the utility of post-marketing AE reporting.

Electronic supplementary material

The online version of this article (doi:10.1007/s40264-014-0150-2) contains supplementary material, which is available to authorized users.

Systems pharmacology augments drug safety surveillance

Small molecule drugs are the foundation of modern medical practice yet their use is limited by the onset of unexpected and severe adverse events (AEs). Regulatory agencies rely on post-marketing surveillance to monitor safety once drugs are approved for clinical use. Despite advances in pharmacovigilance methods that address issues of confounding bias, clinical data of AEs are inherently noisy. Systems pharmacology– the integration of systems biology and chemical genomics – can illuminate drug mechanisms of action. We hypothesize that these data can improve drug safety surveillance by highlighting drugs with a mechanistic connection to the target phenotype (enriching true positives) and filtering those that do not (depleting false positives). We present an algorithm, the modular assembly of drug safety subnetworks (MADSS), to combine systems pharmacology and pharmacovigilance data and significantly improve drug safety monitoring for four clinically relevant adverse drug reactions.

Disease risk factors identified through shared genetic architecture and electronic medical records

Genome-wide association studies have identified genetic variants for thousands of diseases and traits. We evaluated the relationships between specific risk factors (for example, blood cholesterol level) and diseases on the basis of their shared genetic architecture in a comprehensive human disease-single-nucleotide polymorphism association database (VARIMED), analyzing the findings from 8962 published association studies. Similarity between traits and diseases was statistically evaluated on the basis of their association with shared gene variants. We identified 120 disease-trait pairs that were statistically similar, and of these, we tested and validated five previously unknown disease-trait associations by searching electronic medical records (EMRs) from three independent medical centers for evidence of the trait appearing in patients within 1 year of first diagnosis of the disease. We validated that the mean corpuscular volume is elevated before diagnosis of acute lymphoblastic leukemia; both have associated variants in the gene IKZF1. Platelet count is decreased before diagnosis of alcohol dependence; both are associated with variants in the gene C12orf51. Alkaline phosphatase level is elevated in patients with venous thromboembolism; both share variants in ABO. Similarly, we found that prostate-specific antigen and serum magnesium levels were altered before the diagnosis of lung cancer and gastric cancer, respectively. Disease-trait associations identify traits that could serve as future prognostics, if validated through EMR and subsequent prospective trials.

Similarity-based modeling applied to signal detection in pharmacovigilance

One of the main objectives in pharmacovigilance is the detection of adverse drug events (ADEs) through mining of healthcare databases, such as electronic health records or administrative claims data. Although different approaches have been shown to be of great value, research is still focusing on the enhancement of signal detection to gain efficiency in further assessment and follow-up. We applied similarity-based modeling techniques, using 2D and 3D molecular structure, ADE, target, and ATC (anatomical therapeutic chemical) similarity measures, to the candidate associations selected previously in a medication-wide association study for four ADE outcomes. Our results showed an improvement in the precision when we ranked the subset of ADE candidates using similarity scorings. This method is simple, useful to strengthen or prioritize signals generated from healthcare databases, and facilitates ADE detection through the identification of the most similar drugs for which ADE information is available.

Similarity-based modeling in large-scale prediction of drug-drug interactions

Drug-drug interactions (DDIs) are a major cause of adverse drug effects and a public health concern, as they increase hospital care expenses and reduce patients' quality of life. DDI detection is, therefore, an important objective in patient safety, one whose pursuit affects drug development and pharmacovigilance. In this article, we describe a protocol applicable on a large scale to predict novel DDIs based on similarity of drug interaction candidates to drugs involved in established DDIs. The method integrates a reference standard database of known DDIs with drug similarity information extracted from different sources, such as 2D and 3D molecular structure, interaction profile, target and side-effect similarities. The method is interpretable in that it generates drug interaction candidates that are traceable to pharmacological or clinical effects. We describe a protocol with applications in patient safety and preclinical toxicity screening. The time frame to implement this protocol is 5-7 h, with additional time potentially necessary, depending on the complexity of the reference standard DDI database and the similarity measures implemented.

Insight into the interactions between novel coumarin derivatives and human A3 adenosine receptors

A study focused on the discovery of new chemical entities based on the 3-arylcoumarin scaffold was performed with the aim of finding new adenosine receptor (AR) ligands. Thirteen synthesized compounds were evaluated by radioligand binding (A1, A2A, and A3) and adenylyl cyclase activity (A2B) assays in order to study their affinity for the four human AR (hAR) subtypes. Seven of the studied compounds proved to be selective A3 AR ligands, with 3-(4'-methylphenyl)-8-(2-oxopropoxy)coumarin (12) being the most potent (Ki =634 nM). None of the compounds showed affinity for the A2B receptor, while four compounds were found to be nonselective AR ligands for the other three subtypes. Docking simulations were carried out to identify the hypothetical binding mode and to rationalize the interaction of these types of coumarin derivatives with the binding site of the three ARs to which binding was observed. The results allowed us to conclude that the 3-arylcoumarin scaffold composes a novel and promising class of A3 AR ligands. ADME properties were also calculated, with the results suggesting that these compounds are promising leads for the identification of new drug candidates.

Pharmacological inhibition of cystine-glutamate exchange induces endoplasmic reticulum stress and ferroptosis

Exchange of extracellular cystine for intracellular glutamate by the antiporter system x_c⁻ is implicated in numerous pathologies. Pharmacological agents that inhibit system x_c⁻ activity with high potency have long been sought, but have remained elusive. In this study, we report that the small molecule erastin is a potent, selective inhibitor of system x_c⁻. RNA sequencing revealed that inhibition of cystine–glutamate exchange leads to activation of an ER stress response and upregulation of CHAC1, providing a pharmacodynamic marker for system x_c⁻ inhibition. We also found that the clinically approved anti-cancer drug sorafenib, but not other kinase inhibitors, inhibits system x_c⁻ function and can trigger ER stress and ferroptosis. In an analysis of hospital records and adverse event reports, we found that patients treated with sorafenib exhibited unique metabolic and phenotypic alterations compared to patients treated with other kinase-inhibiting drugs. Finally, using a genetic approach, we identified new genes dramatically upregulated in cells resistant to ferroptosis.

DOI:http://dx.doi.org/10.7554/eLife.02523.001

Sugars, fats, amino acids, and other nutrients cannot simply diffuse into the cell. Rather, they must be transported across the cell membrane by specific proteins that stretch from one side of the cell membrane to the other. One such ‘transporter’—system x_c⁻—is of special interest. This transporter imports one molecule of cystine from outside the cell in exchange for one molecule of glutamate from inside the cell. Cystine, a variant of the amino acid cysteine, is essential for synthesizing new proteins and for preventing the accumulation of toxic species inside the cell. Not surprisingly, many cancer cells are dependent upon the transport activity of system x_c⁻ for growth and survival. Drugs that can inhibit system x_c⁻ could therefore be part of potential treatments for cancer and other diseases.

Dixon, Patel, et al. have found that the compound erastin is a very effective inhibitor of system x_c⁻ function. Certain versions of erastin are over 1000 times more potent than the previously known best inhibitor of system x_c⁻, sulfasalazine. Dixon, Patel et al. found that using erastin and sulfasalazine to inhibit system x_c⁻ in cancer cells grown in a petri dish results in an unusual type of iron-dependent cell death called ferroptosis. By inhibiting the uptake of cystine, erastin and other system x_c⁻ inhibitors interfere with the cellular machinery that folds proteins into their final, three-dimensional shape. The accumulation of these partially-folded proteins in the cell causes a specific kind of cellular stress that can be used as a readout, or biomarker, for the inhibition of system x_c⁻. Such a biomarker will be essential for identifying cells in the body that have been exposed to agents that inhibit system x_c⁻ and that are undergoing ferroptosis.

Unexpectedly, Dixon, Patel et al. also found that the FDA-approved anti-cancer drug sorafenib inhibits system x_c⁻. Other drugs in the same class as sorafenib do not share this unusual property. Dixon, Patel, et al. synthesized variants of sorafenib and identified sites on the drug that are necessary for it to be able to interfere with system x_c⁻. Alongside the erastin derivatives, these new molecules may help to develop new drugs that can inhibit this important transporter in a clinical setting.

DOI:http://dx.doi.org/10.7554/eLife.02523.002

Coherent functional modules improve transcription factor target identification, cooperativity prediction, and disease association

Transcription factors (TFs) are fundamental controllers of cellular regulation that function in a complex and combinatorial manner. Accurate identification of a transcription factor's targets is essential to understanding the role that factors play in disease biology. However, due to a high false positive rate, identifying coherent functional target sets is difficult. We have created an improved mapping of targets by integrating ChIP-Seq data with 423 functional modules derived from 9,395 human expression experiments. We identified 5,002 TF-module relationships, significantly improved TF target prediction, and found 30 high-confidence TF-TF associations, of which 14 are known. Importantly, we also connected TFs to diseases through these functional modules and identified 3,859 significant TF-disease relationships. As an example, we found a link between MEF2A and Crohn's disease, which we validated in an independent expression dataset. These results show the power of combining expression data and ChIP-Seq data to remove noise and better extract the associations between TFs, functional modules, and disease.

Transcription factors (TFs) are crucial to the precise regulation of many cellular processes and thus, are responsible for many human phenotypes and diseases. Now that the ENCODE project has mapped hundreds of TFs to their genomic binding locations, extracting functional biological signals is the next step in understanding their role in disease. In this paper, we present a novel approach to identifying TF targets and use these targets to find regulatory relationships between TFs and diseases. We present a large open dataset of putative TF-TF interactions and TF-disease associations which includes known connections as well as novel ones. We validate the association of one of our novel TF-disease associations, MEF2A and Crohn's disease, suggesting that our approach generates testable disease association hypotheses. Integrating these datasets will be crucial for understanding phenotypes and complex diseases.

STORMSeq: an open-source, user-friendly pipeline for processing personal genomics data in the cloud

The increasing public availability of personal complete genome sequencing data has ushered in an era of democratized genomics. However, read mapping and variant calling software is constantly improving and individuals with personal genomic data may prefer to customize and update their variant calls. Here, we describe STORMSeq (Scalable Tools for Open-Source Read Mapping), a graphical interface cloud computing solution that does not require a parallel computing environment or extensive technical experience. This customizable and modular system performs read mapping, read cleaning, and variant calling and annotation. At present, STORMSeq costs approximately $2 and 5–10 hours to process a full exome sequence and $30 and 3–8 days to process a whole genome sequence. We provide this open-access and open-source resource as a user-friendly interface in Amazon EC2.

Genome wide analysis of drug-induced torsades de pointes: lack of common variants with large effect sizes

Marked prolongation of the QT interval on the electrocardiogram associated with the polymorphic ventricular tachycardia Torsades de Pointes is a serious adverse event during treatment with antiarrhythmic drugs and other culprit medications, and is a common cause for drug relabeling and withdrawal. Although clinical risk factors have been identified, the syndrome remains unpredictable in an individual patient. Here we used genome-wide association analysis to search for common predisposing genetic variants. Cases of drug-induced Torsades de Pointes (diTdP), treatment tolerant controls, and general population controls were ascertained across multiple sites using common definitions, and genotyped on the Illumina 610k or 1M-Duo BeadChips. Principal Components Analysis was used to select 216 Northwestern European diTdP cases and 771 ancestry-matched controls, including treatment-tolerant and general population subjects. With these sample sizes, there is 80% power to detect a variant at genome-wide significance with minor allele frequency of 10% and conferring an odds ratio of ≥2.7. Tests of association were carried out for each single nucleotide polymorphism (SNP) by logistic regression adjusting for gender and population structure. No SNP reached genome wide-significance; the variant with the lowest P value was rs2276314, a non-synonymous coding variant in C18orf21 (p  =  3×10⁻⁷, odds ratio = 2, 95% confidence intervals: 1.5–2.6). The haplotype formed by rs2276314 and a second SNP, rs767531, was significantly more frequent in controls than cases (p  =  3×10⁻⁹). Expanding the number of controls and a gene-based analysis did not yield significant associations. This study argues that common genomic variants do not contribute importantly to risk for drug-induced Torsades de Pointes across multiple drugs.

Connecting the dots: applications of network medicine in pharmacology and disease

In 2011, >2.5 million people died from only 15 causes in the United States. Ten of these involved complex or infectious diseases for which there is insufficient knowledge or treatment, such as heart disease, influenza, and Alzheimer's disease.(1) Complex diseases have been difficult to understand due to their multifarious genetic and molecular fingerprints, while certain infectious agents have evolved to elude treatment and prophylaxis. Network medicine provides a macroscopic approach to understanding and treating such illnesses. It integrates experimental data on gene, protein, and metabolic interactions with clinical knowledge of disease and pharmacology in order to extend the understanding of diseases and their treatments. The resulting "big picture" allows for the development of computational and mathematical methods to identify novel disease pathways and predict patient drug response, among others. In this review, we discuss recent advances in network medicine.

Detection of drug-drug interactions by modeling interaction profile fingerprints

Drug-drug interactions (DDIs) constitute an important problem in postmarketing pharmacovigilance and in the development of new drugs. The effectiveness or toxicity of a medication could be affected by the co-administration of other drugs that share pharmacokinetic or pharmacodynamic pathways. For this reason, a great effort is being made to develop new methodologies to detect and assess DDIs. In this article, we present a novel method based on drug interaction profile fingerprints (IPFs) with successful application to DDI detection. IPFs were generated based on the DrugBank database, which provided 9,454 well-established DDIs as a primary source of interaction data. The model uses IPFs to measure the similarity of pairs of drugs and generates new putative DDIs from the non-intersecting interactions of a pair. We described as part of our analysis the pharmacological and biological effects associated with the putative interactions; for example, the interaction between haloperidol and dicyclomine can cause increased risk of psychosis and tardive dyskinesia. First, we evaluated the method through hold-out validation and then by using four independent test sets that did not overlap with DrugBank. Precision for the test sets ranged from 0.4–0.5 with more than two fold enrichment factor enhancement. In conclusion, we demonstrated the usefulness of the method in pharmacovigilance as a DDI predictor, and created a dataset of potential DDIs, highlighting the etiology or pharmacological effect of the DDI, and providing an exploratory tool to facilitate decision support in DDI detection and patient safety.

Web-scale pharmacovigilance: listening to signals from the crowd

Adverse drug events cause substantial morbidity and mortality and are often discovered after a drug comes to market. We hypothesized that Internet users may provide early clues about adverse drug events via their online information-seeking. We conducted a large-scale study of Web search log data gathered during 2010. We pay particular attention to the specific drug pairing of paroxetine and pravastatin, whose interaction was reported to cause hyperglycemia after the time period of the online logs used in the analysis. We also examine sets of drug pairs known to be associated with hyperglycemia and those not associated with hyperglycemia. We find that anonymized signals on drug interactions can be mined from search logs. Compared to analyses of other sources such as electronic health records (EHR), logs are inexpensive to collect and mine. The results demonstrate that logs of the search activities of populations of computer users can contribute to drug safety surveillance.

Novel (coumarin-3-yl)carbamates as selective MAO-B inhibitors: synthesis, in vitro and in vivo assays, theoretical evaluation of ADME properties and docking study

A series of (coumarin-3-yl)carbamates was synthesized and evaluated in vitro as monoamine oxidase (MAO-A and MAO-B) inhibitors. Most of the new compounds selectively inhibited MAO-B isoenzyme with IC50 values in the micro or nanoMolar ranges. Since these compounds must achieve the brain cells, theoretical evaluation of ADME properties were also carried out. Compound 8 (benzyl(coumarin-3-yl)carbamate), which presented the most interesting in vitro MAO-B inhibitory profile (IC50 against MAO-B = 45 nM), was subjected to further studies. This in vitro MAO-B inhibitory activity is comparable with that of the selegiline, the reference compound (IC50 against MAO-B = 20 nM). Taking into account the in vitro results of compound 8, in vivo assays and docking calculations were also carried out for this derivative.

Data-driven prediction of drug effects and interactions

Adverse drug events remain a leading cause of morbidity and mortality around the world. Many adverse events are not detected during clinical trials before a drug receives approval for use in the clinic. Fortunately, as part of postmarketing surveillance, regulatory agencies and other institutions maintain large collections of adverse event reports, and these databases present an opportunity to study drug effects from patient population data. However, confounding factors such as concomitant medications, patient demographics, patient medical histories, and reasons for prescribing a drug often are uncharacterized in spontaneous reporting systems, and these omissions can limit the use of quantitative signal detection methods used in the analysis of such data. Here, we present an adaptive data-driven approach for correcting these factors in cases for which the covariates are unknown or unmeasured and combine this approach with existing methods to improve analyses of drug effects using three test data sets. We also present a comprehensive database of drug effects (Offsides) and a database of drug-drug interaction side effects (Twosides). To demonstrate the biological use of these new resources, we used them to identify drug targets, predict drug indications, and discover drug class interactions. We then corroborated 47 (P < 0.0001) of the drug class interactions using an independent analysis of electronic medical records. Our analysis suggests that combined treatment with selective serotonin reuptake inhibitors and thiazides is associated with significantly increased incidence of prolonged QT intervals. We conclude that confounding effects from covariates in observational clinical data can be controlled in data analyses and thus improve the detection and prediction of adverse drug effects and interactions.

A novel signal detection algorithm for identifying hidden drug-drug interactions in adverse event reports

OBJECTIVE

Adverse drug events (ADEs) are common and account for 770 000 injuries and deaths each year and drug interactions account for as much as 30% of these ADEs. Spontaneous reporting systems routinely collect ADEs from patients on complex combinations of medications and provide an opportunity to discover unexpected drug interactions. Unfortunately, current algorithms for such "signal detection" are limited by underreporting of interactions that are not expected. We present a novel method to identify latent drug interaction signals in the case of underreporting.

MATERIALS AND METHODS

We identified eight clinically significant adverse events. We used the FDA's Adverse Event Reporting System to build profiles for these adverse events based on the side effects of drugs known to produce them. We then looked for pairs of drugs that match these single-drug profiles in order to predict potential interactions. We evaluated these interactions in two independent data sets and also through a retrospective analysis of the Stanford Hospital electronic medical records.

RESULTS

We identified 171 novel drug interactions (for eight adverse event categories) that are significantly enriched for known drug interactions (p=0.0009) and used the electronic medical record for independently testing drug interaction hypotheses using multivariate statistical models with covariates.

CONCLUSION

Our method provides an option for detecting hidden interactions in spontaneous reporting systems by using side effect profiles to infer the presence of unreported adverse events.

An integrative method for scoring candidate genes from association studies: application to warfarin dosing

Background

A key challenge in pharmacogenomics is the identification of genes whose variants contribute to drug response phenotypes, which can include severe adverse effects. Pharmacogenomics GWAS attempt to elucidate genotypes predictive of drug response. However, the size of these studies has severely limited their power and potential application. We propose a novel knowledge integration and SNP aggregation approach for identifying genes impacting drug response. Our SNP aggregation method characterizes the degree to which uncommon alleles of a gene are associated with drug response. We first use pre-existing knowledge sources to rank pharmacogenes by their likelihood to affect drug response. We then define a summary score for each gene based on allele frequencies and train linear and logistic regression classifiers to predict drug response phenotypes.

Results

We applied our method to a published warfarin GWAS data set comprising 181 individuals. We find that our method can increase the power of the GWAS to identify both VKORC1 and CYP2C9 as warfarin pharmacogenes, where the original analysis had only identified VKORC1. Additionally, we find that our method can be used to discriminate between low-dose (AUROC=0.886) and high-dose (AUROC=0.764) responders.

Conclusions

Our method offers a new route for candidate pharmacogene discovery from pharmacogenomics GWAS, and serves as a foundation for future work in methods for predictive pharmacogenomics.

Improving the prediction of pharmacogenes using text-derived drug-gene relationships

A critical goal of pharmacogenomics research is to identify genes that can explain variation in drug response. We have previously reported a method that creates a genome-scale ranking of genes likely to interact with a drug. The algorithm uses information about drug structure and indications of use to rank the genes. Although the algorithm has good performance, its performance depends on a curated set of drug-gene relationships that is expensive to create and difficult to maintain. In this work, we assess the utility of text mining in extracting a network of drug-gene relationships automatically. This provides a valuable aggregate source of knowledge, subsequently used as input into the algorithm that ranks potential pharmacogenes. Using a drug-gene network created from sentence-level co-occurrence in the full text of scientific articles, we compared the performance to that of a network created by manual curation of those articles. Under a wide range of conditions, we show that a knowledge base derived from text-mining the literature performs as well as, and sometimes better than, a high-quality, manually curated knowledge base. We conclude that we can use relationships mined automatically from the literature as a knowledgebase for pharmacogenomics relationships. Additionally, when relationships are missed by text mining, our system can accurately extrapolate new relationships with 77.4% precision.

Predicting drug side-effects by chemical systems biology

New approaches to predicting ligand similarity and protein interactions can explain unexpected observations of drug inefficacy or side-effects.