Search

Single cell technologies are rapidly generating large amounts of data that enables us to understand biological systems at single-cell resolution. However, joint analysis of datasets generated by independent labs remains challenging due to a lack of consistent terminology to describe cell types. Here, we present OnClass, an algorithm and accompanying software for automatically classifying cells into cell types that are part of the controlled vocabulary that forms the Cell Ontology. A key advantage of OnClass is its capability to classify cells into cell types not present in the training data because it uses the Cell Ontology graph to infer cell type relationships. Furthermore, OnClass can be used to identify marker genes for all the cell ontology categories, regardless of whether the cell types are present or absent in the training data, suggesting that OnClass goes beyond a simple annotation tool for single cell datasets, being the first algorithm capable to identify marker genes specific to all terms of the Cell Ontology and offering the possibility of refining the Cell Ontology using a data-centric approach., Classifying cells into unseen cell types remains challenging in scRNA-seq analysis. Here we show that Cell Ontology enables an accurate classification of unseen cell types through considering the cell type relationships in the Cell Ontology graph.

Adverse drug reactions (ADRs) are a major concern for patients, clinicians, and regulatory agencies. The discovery of serious ADRs leading to substantial morbidity and mortality has resulted in mandatory phase IV clinical trials, black box warnings, and withdrawal of drugs from the market. Real-world data, data collected during routine clinical care, is being adopted by innovators, regulators, payors, and providers to inform decision making throughout the product life cycle. We outline several different approaches to modern pharmacovigilance, including spontaneous reporting databases, electronic health record monitoring and research frameworks, social media surveillance, and the use of digital devices. Some of these platforms are well-established while others are still emerging or experimental. We highlight both the potential opportunity, as well as the existing challenges within these pharmacovigilance systems that have already begun to impact the drug development process, as well as the landscape of postmarket drug safety monitoring. Further research and investment into different and complementary pharmacovigilance systems is needed to ensure the continued safety of pharmacotherapy.

Although tremendous effort has been put into cell-type annotation, identification of previously uncharacterized cell types in heterogeneous single-cell RNA-seq data remains a challenge. Here we present MARS, a meta-learning approach for identifying and annotating known as well as new cell types. MARS overcomes the heterogeneity of cell types by transferring latent cell representations across multiple datasets. MARS uses deep learning to learn a cell embedding function as well as a set of landmarks in the cell embedding space. The method has a unique ability to discover cell types that have never been seen before and annotate experiments that are as yet unannotated. We apply MARS to a large mouse cell atlas and show its ability to accurately identify cell types, even when it has never seen them before. Further, MARS automatically generates interpretable names for new cell types by probabilistically defining a cell type in the embedding space.

Asians as a group comprise > 60% the world’s population. There is an incredible amount of diversity in Asian and admixed populations that has not been addressed in a pharmacogenetic context. The known pharmacogenetic differences in Asian subgroups generally represent previously known variants that are present at much lower or higher frequencies in Asians compared with other populations. In this review we summarize the main drugs and known genes that appear to have differences in their pharmacogenetic properties in certain Asian populations. Evidence‐based guidelines and summary statistics from the US Food and Drug Administration and the Clinical Pharmacogenetics Implementation Consortium were analyzed for ethnic differences in outcomes. Implicated drugs included commonly prescribed drugs such as warfarin, clopidogrel, carbamazepine, and allopurinol. The majority of these associations are due to Asians more commonly being poor metabolizers of cytochrome P450 (CYP) 2C19 and carriers of the human leukocyte antigen (HLA)‐B*15:02 allele. The relative risk increase was shown to vary between genes and drugs, but could be > 100‐fold higher in Asians. Specifically, there was a 172‐fold increased risk of Stevens‒Johnson syndrome and toxic epidermal necrolysis with carbamazepine use among HLA‐B*15:02 carriers. The effects ranged from relatively benign reactions such as reduced drug efficacy to severe cutaneous skin reactions. These reactions are severe and prevalent enough to warrant pharmacogenetic testing and appropriate changes in dose and medication choice for at‐risk populations. Further studies should be done on Asian cohorts to more fully understand pharmacogenetic variants in these populations and to clarify how such differences may influence drug response.

The CACNA1S gene encodes the α1s subunit of the dihydropyridine receptor (DHPR), a voltage-gated calcium channel and voltage sensor for Ca(2+) release in skeletal muscle. Variants in the CACNA1S gene have been linked to the pharmacogenetic disorder known as malignant hyperthermia susceptibility (MHS) and hypokalemic periodic paralysis (hypoPP). Two variants in CACNA1S are verified by the European Malignant Hyperthermia Group (EMHG) to be associated with MHS [1]. Although the occurrence of MHS during anesthesia is relatively rare, the genetic prevalence of MHS-causative mutations is estimated to be between 1 in 400 [2] to 1 in 2000–3000 [3]. The American College of Medical Genetics (ACMG) “Guidelines for Reporting Incidental Findings in Clinical Exome and Genome Sequencing” includes “known pathogenic” and “likely pathogenic” variants in CACNA1S related to MHS in its list of genetic variants to report as incidental findings [4, 5]. In addition, 9 CACNA1S variants are linked to hypoPP [6]. Finally, several CACNA1S variants and polymorphisms are proposed to be associated with thyrotoxic periodic paralysis (TPP) [7], hyperCKemia [8] and statin-associated myopathy [9]. Thus, the CACNA1S gene is associated with an eclectic array of muscle disorders with clinical manifestations ranging from subclinical myopathies, episodic paralysis, and life-threatening response to anesthesia.

Objectives We sought to cluster biological phenotypes using semantic similarity and create an easy-to-install, stable, and reproducible tool. Materials and Methods We generated Phenotype Clustering (PhenClust)—a novel application of semantic similarity for interpreting biological phenotype associations—using the Unified Medical Language System (UMLS) metathesaurus, demonstrated the tool’s application, and developed Docker containers with stable installations of two UMLS versions. Results PhenClust identified disease clusters for drug network-associated phenotypes and a meta-analysis of drug target candidates. The Dockerized containers eliminated the requirement that the user install the UMLS metathesaurus. Discussion Clustering phenotypes summarized all phenotypes associated with a drug network and two drug candidates. Docker containers can support dissemination and reproducibility of tools that are otherwise limited due to insufficient software support. Conclusion PhenClust can improve interpretation of high-throughput biological analyses where many phenotypes are associated with a query and the Dockerized PhenClust achieved our objective of decreasing installation complexity.

The 21st Century Cures Act passed by the United States Congress mandates the US Food and Drug Administration to develop guidance to evaluate the use of real‐world evidence (RWE) to support the regulatory process. RWE has generated important medical discoveries, especially in areas where traditional clinical trials would be unethical or infeasible. However, RWE suffers from several issues that hinder its ability to provide proof of treatment efficacy at a level comparable to randomized controlled trials. In this review article, we summarized the advantages and limitations of RWE, identified the key opportunities for RWE, and pointed the way forward to maximize the potential of RWE for regulatory purposes.

Accurate prediction of antigen presentation by human leukocyte antigen (HLA) class II molecules would be valuable for vaccine development and cancer immunotherapies. Current computational methods trained on in vitro binding data are limited by insufficient training data and algorithmic constraints. Here we describe MARIA (major histocompatibility complex analysis with recurrent integrated architecture; https://maria.stanford.edu/ ), a multimodal recurrent neural network for predicting the likelihood of antigen presentation from a gene of interest in the context of specific HLA class II alleles. In addition to in vitro binding measurements, MARIA is trained on peptide HLA ligand sequences identified by mass spectrometry, expression levels of antigen genes and protease cleavage signatures. Because it leverages these diverse training data and our improved machine learning framework, MARIA (area under the curve = 0.89–0.92) outperformed existing methods in validation datasets. Across independent cancer neoantigen studies, peptides with high MARIA scores are more likely to elicit strong CD4+ T cell responses. MARIA allows identification of immunogenic epitopes in diverse cancers and autoimmune disease. A neural network trained on diverse datasets improves prediction of HLA class II epitope presentation.

Accurate determination of target-ligand interactions is crucial in the drug discovery process. In this paper, we propose a two-staged graph-convolutional (Graph-CNN) framework for predicting protein-ligand interactions. We first describe an unsupervised graph-autoencoder to learn fixed-size representations of protein pockets. Two Graph-CNNs are then trained to automatically extract features from pocket graphs and 2D molecular graphs, respectively. We demonstrate that graph-autoencoders can learn meaningful fixed-size representation for protein pockets of varying sizes and the Graph-CNN framework can effectively capture protein-ligand binding interactions without relying on target-ligand co-complexes. Across several metrics, Graph-CNNs achieved better or comparable performance to 3DCNN ligand-scoring, AutoDock Vina, RF-Score, and NNScore on common virtual screening benchmark datasets. Visualization of key pocket residues and ligand atoms contributing to the classification decisions confirms that our networks recognize meaningful interactions between pockets and ligands.Availability and ImplementationContact: rbaltman@stanford.eduSupplementary information

Electronic communication is becoming increasingly popular worldwide, as evidenced by its widespread and rapidly growing use. In medicine, however, it remains a novel approach to reach out to patients. Yet, they have the potential for further improving current health care. Electronic platforms could support therapy adherence and communication between physicians and patients. The power of social media as well as other electronic devices can improve adherence as evidenced by the development of the app bant. Additionally, systemic analysis of social media content by Screenome can identify health events not always captured by regular health care. By better identifying these healthcare events we can improve our current healthcare system as we will be able to better tailor to the patients’ needs. All these techniques are a valuable component of modern health care and will help us into the future of increasingly digital health care.

Aims Clopidogrel is prescribed for the prevention of atherothrombotic events. While investigations have identified genetic determinants of inter-individual variability in on-treatment platelet inhibition (e.g. CYP2C19*2), evidence that these variants have clinical utility to predict major adverse cardiovascular events (CVEs) remains controversial. Methods and results We assessed the impact of 31 candidate gene polymorphisms on adenosine diphosphate (ADP)-stimulated platelet reactivity in 3391 clopidogrel-treated coronary artery disease patients of the International Clopidogrel Pharmacogenomics Consortium (ICPC). The influence of these polymorphisms on CVEs was tested in 2134 ICPC patients (N = 129 events) in whom clinical event data were available. Several variants were associated with on-treatment ADP-stimulated platelet reactivity (CYP2C19*2, P = 8.8 × 10−54; CES1 G143E, P = 1.3 × 10−16; CYP2C19*17, P = 9.5 × 10−10; CYP2B6 1294 + 53 C > T, P = 3.0 × 10−4; CYP2B6 516 G > T, P = 1.0 × 10−3; CYP2C9*2, P = 1.2 × 10−3; and CYP2C9*3, P = 1.5 × 10−3). While no individual variant was associated with CVEs, generation of a pharmacogenomic polygenic response score (PgxRS) revealed that patients who carried a greater number of alleles that associated with increased on-treatment platelet reactivity were more likely to experience CVEs (β = 0.17, SE 0.06, P = 0.01) and cardiovascular-related death (β = 0.43, SE 0.16, P = 0.007). Patients who carried eight or more risk alleles were significantly more likely to experience CVEs [odds ratio (OR) = 1.78, 95% confidence interval (CI) 1.14–2.76, P = 0.01] and cardiovascular death (OR = 4.39, 95% CI 1.35–14.27, P = 0.01) compared to patients who carried six or fewer of these alleles. Conclusion Several polymorphisms impact clopidogrel response and PgxRS is a predictor of cardiovascular outcomes. Additional investigations that identify novel determinants of clopidogrel response and validating polygenic models may facilitate future precision medicine strategies.

The CYP4F2 gene is known to influence mean coumarin dose. The aim of the present study was to undertake a meta-analysis at individual patients' level to capture the possible effect of ethnicity, gene-gene interaction or other drugs on the association and to verify if inclusion of CYP4F2*3 variant into dosing algorithms improves the prediction of mean coumarin dose. We asked the authors of our previous meta-analysis (30 articles) and of 38 new articles retrieved by a systematic review to send us individual patients' data. The final collection consists 15,754 patients split into a derivation and validation cohort. The CYP4F2*3 polymorphism was consistently associated with an increase in mean coumarin dose (+9% (95%CI 7-10%), with a higher effect in females, in patients taking acenocoumarol and in Whites. The inclusion of the CYP4F2*3 in dosing algorithms slightly improved the prediction of stable coumarin dose. New pharmacogenetic equations potentially useful for clinical practice were derived. This article is protected by copyright. All rights reserved.

Pharmacogenomics aims to associate human genetic variability with differences in drug phenotypes in order to tailor drug treatment to individual patients. The massive amount of genetic data generated from large cohorts of patients with variable drug phenotypes have led to advances in this field. Understanding the application of pharmacogenomics in dermatology could inform clinical practice and provide insight for future research. The Pharmacogenomics Knowledge Base and the Clinical Pharmacogenetics Implementation Consortium are among the resources to help clinicians and researchers navigate the many gene-drug associations that have already been discovered. The implementation of clinical pharmacogenomics within health care systems remains an area of ongoing development. This review provides an introduction to the field of pharmacogenomics and to current pharmacogenomics resources using examples of gene-drug associations relevant to the field of dermatology.

Selective serotonin reuptake inhibitors (SSRIs) are first-line antidepressants for the treatment of major depressive disorder (MDD). However, treatment response during an initial therapeutic trial is often poor and is difficult to predict. Heterogeneity of response to SSRIs in depressed patients is partly driven by co-occurring somatic disorders such as coronary artery disease (CAD) and obesity. CAD and obesity may also be associated with metabolic side effects of SSRIs. In this study, we assessed the association of CAD and obesity with treatment response to SSRIs in patients with MDD using a polygenic score (PGS) approach. Additionally, we performed cross-trait meta-analyses to pinpoint genetic variants underpinnings the relationship of CAD and obesity with SSRIs treatment response. First, PGSs were calculated at different p value thresholds (PT) for obesity and CAD. Next, binary logistic regression was applied to evaluate the association of the PGSs to SSRIs treatment response in a discovery sample (ISPC, N = 865), and in a replication cohort (STAR*D, N = 1,878). Finally, a cross-trait GWAS meta-analysis was performed by combining summary statistics. We show that the PGSs for CAD and obesity were inversely associated with SSRIs treatment response. At the most significant thresholds, the PGS for CAD and body mass index accounted 1.3%, and 0.8% of the observed variability in treatment response to SSRIs, respectively. In the cross-trait meta-analyses, we identified (1) 14 genetic loci (including NEGR1, CADM2, PMAIP1, PARK2) that are associated with both obesity and SSRIs treatment response; (2) five genetic loci (LINC01412, PHACTR1, CDKN2B, ATXN2, KCNE2) with effects on CAD and SSRIs treatment response. Our findings implicate that the genetic variants of CAD and obesity are linked to SSRIs treatment response in MDD. A better SSRIs treatment response might be achieved through a stratified allocation of treatment for MDD patients with a genetic risk for obesity or CAD.

Background Defining clinical phenotypes provides opportunities for new diagnostics and may provide insights into early intervention and disease prevention. There is increasing evidence that patient-derived health data may contain information that complements traditional methods of clinical phenotyping. The utility of these data for defining meaningful phenotypic groups is of great interest because social media and online resources make it possible to query large cohorts of patients with health conditions. Methods We evaluated the degree to which patient-reported categorical data is useful for discovering subclinical phenotypes and evaluated its utility for discovering new measures of disease severity, treatment response and genetic architecture. Specifically, we examined the responses of 1961 patients with inflammatory bowel disease to questionnaires in search of sub-phenotypes. We applied machine learning methods to identify novel subtypes of Crohn’s disease and studied their associations with drug responses. Results Using the patients’ self-reported information, we identified two subpopulations of Crohn’s disease; these subpopulations differ in disease severity, associations with smoking, and genetic transmission patterns. We also identified distinct features of drug response for the two Crohn’s disease subtypes. These subtypes show a trend towards differential genotype signatures. Conclusion Our findings suggest that patient-defined data can have unplanned utility for defining disease subtypes and may be useful for guiding treatment approaches.

Adverse drug reactions (ADRs) impact the health of 100,000s of individuals annually in the United States with associated costs in the hundreds of billions. The monitoring and analysis of the severity of adverse drug reactions is limited by the current qualitative and categorical system of severity classifications. Previous efforts have generated quantitative estimates for a subset of ADRs, but were limited in scope due to the time and costs associated with the efforts. We present a semi-supervised approach that estimates ADR severity by using a lexical network of ADR word embeddings and label propagation. We use this method to estimate the severity of 28,113 ADRs, representing 12,198 unique ADR concepts from MedDRA. Our Severity of Adverse Events Derived from Reddit (Saedr) scores have good correlations with real-world outcomes. Saedr scores had Spearman correlations with ADR case outcomes in FAERS of 0.595, 0.633, and −0.748 for death, serious outcome, and no outcome, respectively. We investigate different methods for defining initial seed term sets and evaluate their impact on severity estimates. We analyzed severity distributions for ADRs based on their appearance in Boxed Warning drug label sections, as well as ADRs with sex-specific associations. We find that ADRs discovered postmarket have significantly greater severity compared to those discovered in the clinical trial. We create quantitative Drug RIsk Profile (Drip) scores for 968 drugs that have a Spearman correlation of 0.377 with drugs ranked by FAERS cases resulting in death, where the given drug was the primary suspect. We make the Saedr and Drip scores publicly available in order to enable more quantitative analysis of pharmacovigilance data.

Background Adverse drug reactions (ADRs) affect the health of hundreds of thousands of individuals annually in the United States, with associated costs of hundreds of billions of dollars. The monitoring and analysis of the severity of ADRs is limited by the current qualitative and categorical systems of severity classification. Previous efforts have generated quantitative estimates for a subset of ADRs but were limited in scope because of the time and costs associated with the efforts. Objective The aim of this study is to increase the number of ADRs for which there are quantitative severity estimates while improving the quality of these severity estimates. Methods We present a semisupervised approach that estimates ADR severity by using social media word embeddings to construct a lexical network of ADRs and perform label propagation. We used this method to estimate the severity of 28,113 ADRs, representing 12,198 unique ADR concepts from the Medical Dictionary for Regulatory Activities. Results Our Severity of Adverse Events Derived from Reddit (SAEDR) scores have good correlations with real-world outcomes. The SAEDR scores had Spearman correlations of 0.595, 0.633, and −0.748 for death, serious outcome, and no outcome, respectively, with ADR case outcomes in the Food and Drug Administration Adverse Event Reporting System. We investigated different methods for defining initial seed term sets and evaluated their impact on the severity estimates. We analyzed severity distributions for ADRs based on their appearance in boxed warning drug label sections, as well as for ADRs with sex-specific associations. We found that ADRs discovered in the postmarketing period had significantly greater severity than those discovered during the clinical trial (P Conclusions Our SAEDR and DRIP scores are well correlated with the real-world outcomes of the entities they represent and have demonstrated utility in pharmacovigilance research. We make the SAEDR scores for 12,198 ADRs and the DRIP scores for 968 drugs publicly available to enable more quantitative analysis of pharmacovigilance data.

For many prevalent complex diseases, treatment regimens are frequently ineffective. For example, despite multiple available immunomodulators and immunosuppressants, inflammatory bowel disease (IBD) remains difficult to treat. Heterogeneity in the disease across patients makes it challenging to select the optimal treatment regimens, and some patients do not respond to any of the existing treatment choices. Drug repurposing strategies for IBD have had limited clinical success and have not typically offered individualized patient-level treatment recommendations. In this work, we present NetPTP, a Network-based Personalized Treatment Prediction framework which models measured drug effects from gene expression data and applies them to patient samples to generate personalized ranked treatment lists. To accomplish this, we combine publicly available network, drug target, and drug effect data to generate treatment rankings using patient data. These ranked lists can then be used to prioritize existing treatments and discover new therapies for individual patients. We demonstrate how NetPTP captures and models drug effects, and we apply our framework to individual IBD samples to provide novel insights into IBD treatment., Author summary Offering personalized treatment results is an important tenant of precision medicine, particularly in complex diseases which have high variability in disease manifestation and treatment response. We have developed a novel framework, NetPTP (Network-based Personalized Treatment Prediction), for making personalized drug ranking lists for patient samples. Our method uses networks to model drug effects from gene expression data and applies these captured effects to individual samples to produce tailored drug treatment rankings. We applied NetPTP to inflammatory bowel disease, yielding insights into the treatment of this particular disease. Our method is modular and generalizable, and thus can be applied to other diseases that could benefit from a personalized treatment approach.

Pharmacogenetics (PGx) studies the influence of genetic variation on drug response. Clinically actionable associations inform guidelines created by the Clinical Pharmacogenetics Implementation Consortium (CPIC), but the broad impact of genetic variation on entire populations is not well-understood. We analyzed PGx allele and phenotype frequencies for 487,409 participants in the U.K. Biobank, the largest PGx study to date. For fourteen CPIC pharmacogenes known to influence human drug response, we find that 99.5% of individuals may have an atypical response to at least one drug; on average they may have an atypical response to 10.3 drugs. Nearly 24% of participants have been prescribed a drug for which they are predicted to have an atypical response. Non-European populations carry a greater frequency of variants that are predicted to be functionally deleterious; many of these are not captured by current PGx allele definitions. Strategies for detecting and interpreting rare variation will be critical for enabling broad application of pharmacogenetics.

The scale and speed of the COVID-19 pandemic has strained many parts of the national healthcare infrastructure, including communicable disease monitoring and prevention. Many local health departments now receive hundreds or thousands of COVID-19 case reports a day. Many arrive via faxed handwritten forms, often intermingled with other faxes sent to a general fax line, making it difficult to rapidly identify the highest priority cases for outreach and monitoring. We present an AI-based system capable of real-time identification and triage of handwritten faxed COVID-19 forms. The system relies on two models: one model to identify which received pages correspond to case report forms, and a second model to extract information from the set of identified case reports. We evaluated the system on a set of 1,224 faxes received by a local health department over a two-week period. For the 88% of faxes of sufficient quality, the system detects COVID-19 reports with high precision, 0.98, and high recall, 0.91. Among all received COVID-19 faxes, the system identifies high priority cases with a specificity of 0.87, a precision of 0.46 and recall of 0.83. Our system can be adapted to new forms, after a brief training period. Covid Fast Fax can support local health departments in their efforts to control the spread of COVID-19 and limit its impact on the community. The tool is freely available.

Massively accumulated pharmacogenomics, chemogenomics, and side effect datasets offer an unprecedented opportunity for drug response prediction, drug target identification and drug side effect prediction. Existing computational approaches limit their scope to only one of these three tasks, inevitably overlooking the rich connection among them. Here, we propose DrugOrchestra, a deep multi-task learning framework that jointly predicts drug response, targets and side effects. DrugOrchestra leverages pre-trained molecular structure-based drug representation to bridge these three tasks. Instead of directly fine-tuning on an individual task, DrugOrchestra uses deep multi-task learning to obtain a phenotype-based drug representation by simultaneously fine-tuning on drug response, target and side effect prediction. By coupling these three tasks together, DrugOrchestra is able to make predictions for unseen drugs by only knowing their molecular structures. We constructed a heterogeneous drug discovery dataset of over 21k drugs by integrating 8 datasets across three tasks. Our method obtained significant improvement in comparison to methods that were trained on a single task or a single dataset. We further revealed the transferability across 8 datasets and 3 tasks, providing novel insights for understanding drug mechanisms.Availabilityhttps://github.com/jiangdada1221/DrugOrchestra

Background Women are at more than 1.5-fold higher risk for clinically relevant adverse drug events. While this higher prevalence is partially due to gender-related effects, biological sex differences likely also impact drug response. Publicly available gene expression databases provide a unique opportunity for examining drug response at a cellular level. However, missingness and heterogeneity of metadata prevent large-scale identification of drug exposure studies and limit assessments of sex bias. To address this, we trained organism-specific models to infer sample sex from gene expression data, and used entity normalization to map metadata cell line and drug mentions to existing ontologies. Using this method, we inferred sex labels for 450,371 human and 245,107 mouse microarray and RNA-seq samples from refine.bio. Results Overall, we find slight female bias (52.1%) in human samples and (62.5%) male bias in mouse samples; this corresponds to a majority of mixed sex studies in humans and single sex studies in mice, split between female-only and male-only (25.8% vs. 18.9% in human and 21.6% vs. 31.1% in mouse, respectively). In drug studies, we find limited evidence for sex-sampling bias overall; however, specific categories of drugs, including human cancer and mouse nervous system drugs, are enriched in female-only and male-only studies, respectively. We leverage our expression-based sex labels to further examine the complexity of cell line sex and assess the frequency of metadata sex label misannotations (2–5%). Conclusions Our results demonstrate limited overall sex bias, while highlighting high bias in specific subfields and underscoring the importance of including sex labels to better understand the underlying biology. We make our inferred and normalized labels, along with flags for misannotated samples, publicly available to catalyze the routine use of sex as a study variable in future analyses.

Genetics plays a key role in drug response, affecting efficacy and toxicity. Pharmacogenomics aims to understand how genetic variation influences drug response and develop clinical guidelines to aid clinicians in personalized treatment decisions informed by genetics. Although pharmacogenomics has not been broadly adopted into clinical practice, genetics influences treatment decisions regardless. Physicians adjust patient care based on observed response to medication, which may occur as a result of genetic variants harbored by the patient. Here we seek to understand the genetics of drug selection in statin therapy, a class of drugs widely used for high cholesterol treatment. Genetics are known to play an important role in statin efficacy and toxicity, leading to significant changes in patient outcome. We performed genome-wide association studies (GWAS) on statin selection among 59,198 participants in the UK Biobank and found that variants known to influence statin efficacy are significantly associated with statin selection. Specifically, we find that carriers of variants inAPOEandLPAthat are known to decrease efficacy of treatment are more likely to be on atorvastatin, a stronger statin. Additionally, carriers of theAPOEandLPAvariants are more likely to be on a higher intensity dose (a dose that reduces low-density lipoprotein cholesterol by greater than 40%) of atorvastatin than non-carriers (APOE:p(high intensity)= 0.16, OR = 1.7,P= 1.64 × 10−4,LPA:p(high intensity)= 0.17, OR = 1.4,P= 1.14 × 10−2). These findings represent the largest genetic association study of statin selection and statin dose association to date and provide evidence for the role ofLPAandAPOEin statin response, furthering the possibility of personalized statin therapy.

Pharmacogenetics studies how genetic variation leads to variability in drug response. Guidelines for selecting the right drug and right dose to patients based on their genetics are clinically effective, but are still widely unused. For some drugs, the normal clinical decision making process may lead to the optimal dose of a drug that minimizes side effects and maximizes effectiveness. Without measurements of genotype, physicians and patients may observe and adjust dosage in a manner that reflects the underlying genetics. The emergence of genetic data linked to longitudinal clinical data in large biobanks offers an opportunity to confirm known pharmacogenetic interactions as well as discover novel associations by investigating outcomes from normal clinical practice. Here we use the UK Biobank to search for pharmacogenetic interactions among 200 drugs and 9 genes among 200,000 participants. We identify associations between pharmacogene phenotypes and drug maintenance dose as well as side effect incidence. We find support for several known drug-gene associations as well as novel pharmacogenetic interactions.

Requiring regional or in-country confirmatory clinical trials before approval of drugs already approved elsewhere delays access to medicines in low- and middle-income countries and raises drug costs. Here, we discuss the scientific and technological advances that may reduce the need for in-country or in-region clinical trials for drugs approved in other countries and limitations of these advances that could necessitate in-region clinical studies.

Background Enzymatic and chemical reactions are key for understanding biological processes in cells. Curated databases of chemical reactions exist but these databases struggle to keep up with the exponential growth of the biomedical literature. Conventional text mining pipelines provide tools to automatically extract entities and relationships from the scientific literature, and partially replace expert curation, but such machine learning frameworks often require a large amount of labeled training data and thus lack scalability for both larger document corpora and new relationship types. Results We developed an application of Snorkel, a weakly supervised learning framework, for extracting chemical reaction relationships from biomedical literature abstracts. For this work, we defined a chemical reaction relationship as the transformation of chemical A to chemical B. We built and evaluated our system on small annotated sets of chemical reaction relationships from two corpora: curated bacteria-related abstracts from the MetaCyc database (MetaCyc_Corpus) and a more general set of abstracts annotated with MeSH (Medical Subject Headings) term Bacteria (Bacteria_Corpus; a superset of MetaCyc_Corpus). For the MetaCyc_Corpus, we obtained 84% precision and 41% recall (55% F1 score). Extending to the more general Bacteria_Corpus decreased precision to 62% with only a four-point drop in recall to 37% (46% F1 score). Overall, the Bacteria_Corpus contained two orders of magnitude more candidate chemical reaction relationships (nine million candidates vs 68,0000 candidates) and had a larger class imbalance (2.5% positives vs 5% positives) as compared to the MetaCyc_Corpus. In total, we extracted 6871 chemical reaction relationships from nine million candidates in the Bacteria_Corpus. Conclusions With this work, we built a database of chemical reaction relationships from almost 900,000 scientific abstracts without a large training set of labeled annotations. Further, we showed the generalizability of our initial application built on MetaCyc documents enriched with chemical reactions to a general set of articles related to bacteria.

Objective To assess usability and usefulness of a machine learning-based order recommender system applied to simulated clinical cases. Materials and Methods 43 physicians entered orders for 5 simulated clinical cases using a clinical order entry interface with or without access to a previously developed automated order recommender system. Cases were randomly allocated to the recommender system in a 3:2 ratio. A panel of clinicians scored whether the orders placed were clinically appropriate. Our primary outcome included the difference in clinical appropriateness scores. Secondary outcomes included total number of orders, case time, and survey responses. Results Clinical appropriateness scores per order were comparable for cases randomized to the order recommender system (mean difference -0.11 order per score, 95% CI: [-0.41, 0.20]). Physicians using the recommender placed more orders (median 16 vs 15 orders, incidence rate ratio 1.09, 95%CI: [1.01-1.17]). Case times were comparable with the recommender system. Order suggestions generated from the recommender system were more likely to match physician needs than standard manual search options. Physicians used recommender suggestions in 98% of available cases. Approximately 95% of participants agreed the system would be useful for their workflows. Discussion User testing with a simulated electronic medical record interface can assess the value of machine learning and clinical decision support tools for clinician usability and acceptance before live deployments. Conclusions Clinicians can use and accept machine learned clinical order recommendations integrated into an electronic order entry interface in a simulated setting. The clinical appropriateness of orders entered was comparable even when supported by automated recommendations.

ObjectiveTo determine whether clinicians will use machine learned clinical order recommender systems for electronic order entry for simulated inpatient cases, and whether such recommendations impact the clinical appropriateness of the orders being placed.Materials and Methods43 physicians used a clinical order entry interface for five simulated medical cases, with each physician-case randomized whether to have access to a previously-developed clinical order recommendation system. A panel of clinicians determined whether orders placed were clinically appropriate. The primary outcome was the difference in clinical appropriateness scores of orders for cases randomized to the recommender system. Secondary outcomes included usage metrics and physician opinions.ResultsClinical appropriateness scores for orders were comparable for cases randomized to the recommender system (mean difference -0.1 order per score, 95% CI:[-0.4, 0.2]). Physicians using the recommender placed more orders (mean 17.3 vs. 15.7 orders; incidence ratio 1.09, 95% CI:[1.01-1.17]). Case times were comparable with the recommender system. Order suggestions generated from the recommender system were more likely to match physician needs than standard manual search options. Approximately 95% of participants agreed the system would be useful for their workflows.DiscussionMachine-learned clinical order options can meet physician needs better than standard manual search systems. This may increase the number of clinical orders placed per case, while still resulting in similar overall clinically appropriate choices.ConclusionsClinicians can use and accept machine learned clinical order recommendations integrated into an electronic order entry interface. The clinical appropriateness of orders entered was comparable even when supported by automated recommendations.

A BSTRACT The primary challenge of fixed-backbone protein sequence design is to find a distribution of sequences that fold to the backbone of interest. In practice, state-of-the-art protocols often find viable but highly convergent solutions. In this study, we propose a novel method for fixed-backbone protein sequence design using a learned deep neural network potential. We train a convolutional neural network (CNN) to predict a distribution over amino acids at each residue position conditioned on the local structural environment around the residues. Our method for sequence design involves iteratively sampling from this conditional distribution. We demonstrate that this approach is able to produce feasible, novel designs with quality on par with the state-of-the-art, while achieving greater design diversity. In terms of generalizability, our method produces plausible and variable designs for a de novo TIM-barrel structure, showcasing its practical utility in design applications for which there are no known native structures.

Antiplatelet response to clopidogrel shows wide variation, and poor response is correlated with adverse clinical outcomes. CYP2C19 loss-of-function alleles play an important role in this response, but account for only a small proportion of variability in response to clopidogrel. An aim of the International Clopidogrel Pharmacogenomics Consortium (ICPC) is to identify other genetic determinants of clopidogrel pharmacodynamics and clinical response. A genomewide association study (GWAS) was performed using DNA from 2,750 European ancestry individuals, using adenosine diphosphate-induced platelet reactivity and major cardiovascular and cerebrovascular events as outcome parameters. GWAS for platelet reactivity revealed a strong signal for CYP2C19*2 (P value = 1.67e−33). After correction for CYP2C19*2 no other single-nucleotide polymorphism reached genomewide significance. GWAS for a combined clinical end point of cardiovascular death, myocardial infarction, or stroke (5.0% event rate), or a combined end point of cardiovascular death or myocardial infarction (4.7% event rate) showed no significant results, although in coronary artery disease, percutaneous coronary intervention, and acute coronary syndrome subgroups, mutations in SCOS5P1, CDC42BPA, and CTRAC1 showed genomewide significance (lowest P values: 1.07e−09, 4.53e−08, and 2.60e−10, respectively). CYP2C19*2 is the strongest genetic determinant of on-clopidogrel platelet reactivity. We identified three novel associations in clinical outcome subgroups, suggestive for each of these outcomes.

Motivation Accurate annotation of protein functions is fundamental for understanding molecular and cellular physiology. Data-driven methods hold promise for systematically deriving rules underlying the relationship between protein structure and function. However, the choice of protein structural representation is critical. Pre-defined biochemical features emphasize certain aspects of protein properties while ignoring others, and therefore may fail to capture critical information in complex protein sites. Results In this paper, we present a general framework that applies 3D convolutional neural networks (3DCNNs) to structure-based protein functional site detection. The framework can extract task-dependent features automatically from the raw atom distributions. We benchmarked our method against other methods and demonstrate better or comparable performance for site detection. Our deep 3DCNNs achieved an average recall of 0.955 at a precision threshold of 0.99 on PROSITE families, detected 98.89 and 92.88% of nitric oxide synthase and TRYPSIN-like enzyme sites in Catalytic Site Atlas, and showed good performance on challenging cases where sequence motifs are absent but a function is known to exist. Finally, we inspected the individual contributions of each atom to the classification decisions and show that our models successfully recapitulate known 3D features within protein functional sites. Availability and implementation The 3DCNN models described in this paper are available at https://simtk.org/projects/fscnn. Supplementary information Supplementary data are available at Bioinformatics online.

Sertraline, a selective serotonin reuptake inhibitor (SSRI), is commonly prescribed to treat several psychiatric disorders including major depressive disorder, panic, generalized and social anxiety disorders as well as obsessive-compulsive disorder (OCD) [1]. It has similar efficacy to other SSRIs and is considered to cause fewer side effects than some antidepressants, such as tricyclic antidepressants [2, 3]. However, sertraline-induced adverse events such as reversible hepatic injury and Stevens Johnson Syndrome/toxic epidermal necrolysis (SJS/TEN) have been reported [4, 5]. Sertraline is a secondary amine with two chiral centers [6–9]. The potent cis-(1S,4S) enantiomer is used as an antidepressant in either a tablet or an oral solution [1, 7, 10], although the (1R,4R) enantiomer also inhibits serotonin reuptake [10].

One in five Americans experience mental illness, and roughly 75% of psychiatric prescriptions do not successfully treat the patient’s condition. Extensive evidence implicates genetic factors and signaling disruption in the pathophysiology of these diseases. Changes in transcription often underlie this molecular pathway dysregulation; individual patient transcriptional data can improve the efficacy of diagnosis and treatment. Recent large-scale genomic studies have uncovered shared genetic modules across multiple psychiatric disorders—providing an opportunity for an integrated multi-disease approach for diagnosis. Moreover, network-based models informed by gene expression can represent pathological biological mechanisms and suggest new genes for diagnosis and treatment. Here, we use patient gene expression data from multiple studies to classify psychiatric diseases, integrate knowledge from expert-curated databases and publicly available experimental data to create augmented disease-specific gene sets, and use these to recommend disease-relevant drugs. From Gene Expression Omnibus, we extract expression data from 145 cases of schizophrenia, 82 cases of bipolar disorder, 190 cases of major depressive disorder, and 307 shared controls. We use pathway-based approaches to predict psychiatric disease diagnosis with a random forest model (78% accuracy) and derive important features to augment available drug and disease signatures. Using protein-protein-interaction networks and embedding-based methods, we build a pipeline to prioritize treatments for psychiatric diseases that achieves a 3.4-fold improvement over a background model. Thus, we demonstrate that gene-expression-derived pathway features can diagnose psychiatric diseases and that molecular insights derived from this classification task can inform treatment prioritization for psychiatric diseases.

Chemoinformatics is an established discipline focusing on extracting, processing and extrapolating meaningful data from chemical structures. With the rapid explosion of chemical 'big' data from HTS and combinatorial synthesis, machine learning has become an indispensable tool for drug designers to mine chemical information from large compound databases to design drugs with important biological properties. To process the chemical data, we first reviewed multiple processing layers in the chemoinformatics pipeline followed by the introduction of commonly used machine learning models in drug discovery and QSAR analysis. Here, we present basic principles and recent case studies to demonstrate the utility of machine learning techniques in chemoinformatics analyses; and we discuss limitations and future directions to guide further development in this evolving field.

Background Monogenic diseases have been shown to contribute to complex disease risk and may hold new insights into the underlying biological mechanism of Inflammatory Bowel Disease (IBD). Methods We analyzed Mendelian disease associations with IBD using over 55 million patients from the Optum's deidentified electronic health records dataset database. Using the significant Mendelian diseases, we performed pathway enrichment analysis and constructed a model using gene expression datasets to differentiate Crohn's disease (CD), ulcerative colitis (UC), and healthy patient samples. Results We found 50 Mendelian diseases were significantly associated with IBD, with 40 being significantly associated with both CD and UC. Our results for CD replicated those from previous studies. Pathways that were enriched consisted of mainly immune and metabolic processes with a focus on tolerance and oxidative stress. Our 3-way classifier for UC, CD, and healthy samples yielded an accuracy of 72%. Conclusions Mendelian diseases that are significantly associated with IBD may reveal novel insights into the genetic architecture of IBD.

Summary Limited efficacy and intolerable safety limit therapeutic development and identification of potential liabilities earlier in development could significantly improve this process. Computational approaches which aggregate data from multiple sources and consider the drug’s pathways effects could add to identification of these liabilities earlier. Such computational methods must be accessible to a variety of users beyond computational scientists, especially regulators and industry scientists, in order to impact the therapeutic development process. We have previously developed and published PathFX, an algorithm for identifying drug networks and phenotypes for understanding drug associations to safety and efficacy. Here we present a streamlined and easy-to-use PathFX web application that allows users to search for drug networks and associated phenotypes. We have also added visualization, and phenotype clustering to improve functionality and interpretability of PathFXweb. Availability and implementation https://www.pathfxweb.net/. Supplementary information Supplementary data are available at Bioinformatics online.

Linear models offer a robust, flexible, and computationally efficient set of tools for modeling quantitative structure activity relationships (QSAR), but have been eclipsed in performance by non-linear methods. Support vector machines (SVMs) and neural networks are currently among the most popular and accurate QSAR methods because they learn new representations of the data that greatly improve modelability. In this work we use shallow representation learning to improve the accuracy of L1 regularized logistic regression (LASSO) and meet the performance of Tanimoto SVM. We embedded chemical fingerprints in Euclidean space using Tanimoto (aka Jaccard) similarity kernel principal components analysis (KPCA), and compared the effects on LASSO and SVM model performance for predicting the binding activities of chemical compounds against 102 virtual screening targets. We observed similar performance and patterns of improvement for LASSO and SVM. We also empirically measured model training and cross validation times to show that KPCA used in concert with LASSO classification is significantly faster than linear SVM over a wide range of training set sizes. Our work shows that powerful linear QSAR methods can match nonlinear methods, and demonstrates a modular approach to non-linear classification that greatly enhances QSAR model prototyping facility, flexibility, and transferability.

Numerous pharmacogenetic clinical guidelines and recommendations have been published, but barriers have hindered the clinical implementation of pharmacogenetics. The Translational Pharmacogenetics Program (TPP) of the National Institutes of Health (NIH) Pharmacogenomics Research Network was established in 2011 to catalog and contribute to the development of pharmacogenetic implementations at eight US healthcare systems, with the goal to disseminate real-world solutions for the barriers to clinical pharmacogenetic implementation. The TPP collected and normalized pharmacogenetic implementation metrics through June 2015, including gene-drug pairs implemented, interpretations of alleles and diplotypes, numbers of tests performed and actionable results, and workflow diagrams. TPP participant institutions developed diverse solutions to overcome many barriers, but the use of Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines provided some consistency among the institutions. The TPP also collected some pharmacogenetic implementation outcomes (scientific, educational, financial, and informatics), which may inform healthcare systems seeking to implement their own pharmacogenetic testing programs.

Background Central to protein biology is the understanding of how structural elements give rise to observed function. The surfeit of protein structural data enables development of computational methods to systematically derive rules governing structural-functional relationships. However, performance of these methods depends critically on the choice of protein structural representation. Most current methods rely on features that are manually selected based on knowledge about protein structures. These are often general-purpose but not optimized for the specific application of interest. In this paper, we present a general framework that applies 3D convolutional neural network (3DCNN) technology to structure-based protein analysis. The framework automatically extracts task-specific features from the raw atom distribution, driven by supervised labels. As a pilot study, we use our network to analyze local protein microenvironments surrounding the 20 amino acids, and predict the amino acids most compatible with environments within a protein structure. To further validate the power of our method, we construct two amino acid substitution matrices from the prediction statistics and use them to predict effects of mutations in T4 lysozyme structures. Results Our deep 3DCNN achieves a two-fold increase in prediction accuracy compared to models that employ conventional hand-engineered features and successfully recapitulates known information about similar and different microenvironments. Models built from our predictions and substitution matrices achieve an 85% accuracy predicting outcomes of the T4 lysozyme mutation variants. Our substitution matrices contain rich information relevant to mutation analysis compared to well-established substitution matrices. Finally, we present a visualization method to inspect the individual contributions of each atom to the classification decisions. Conclusions End-to-end trained deep learning networks consistently outperform methods using hand-engineered features, suggesting that the 3DCNN framework is well suited for analysis of protein microenvironments and may be useful for other protein structural analyses. Electronic supplementary material The online version of this article (doi:10.1186/s12859-017-1702-0) contains supplementary material, which is available to authorized users.

Objective Determine how varying longitudinal historical training data can impact prediction of future clinical decisions. Estimate the " decay rate " of clinical data source relevance. Materials and methods We trained a clinical order recommender system, analogous to Netflix or Amazon's " Customers who bought A also bought B... " product recommenders, based on a tertiary academic hospital's structured electronic health record data. We used this system to predict future (2013) admission orders based on different subsets of historical training data (2009 through 2012), relative to existing human-authored order sets. Results Predicting future (2013) inpatient orders is more accurate with models trained on just one month of recent (2012) data than with 12 months of older (2009) data (ROC AUC 0.91 vs. 0.88, precision 27% vs. 22%, recall 52% vs. 43%, all P −10 ). Algorithmically learned models from even the older (2009) data was still more effective than existing human-authored order sets (ROC AUC 0.81, precision 16% recall 35%). Training with more longitudinal data (2009–2012) was no better than using only the most recent (2012) data, unless applying a decaying weighting scheme with a " half-life " of data relevance about 4 months. Discussion Clinical practice patterns (automatically) learned from electronic health record data can vary substantially across years. Gold standards for clinical decision support are elusive moving targets, reinforcing the need for automated methods that can adapt to evolving information. Conclusions and relevancm Prioritizing small amounts of recent data is more effective than using larger amounts of older data towards future clinical predictions.

Gene functional enrichment is a mainstay of genomics, but it relies on manually curated databases of gene functions that are incomplete and unaware of the biological context. Here we present an alternative machine learning approach, Deep Functional Synthesis (DeepSyn), which moves beyond gene function databases to dynamically infer the functions of a gene set from its associated network of literature and data, conditioned on the disease and drug context of the current experiment. Using a knowledge graph with 3,048,803 associations between genes, diseases, drugs, and functions, DeepSyn obtained accurate performance (range 0.74 AUC to 0.96 AUC) on a variety of biological applications including drug target identification, gene set functional enrichment, and disease gene prediction.AvailabilityThe DeepSyn codebase is available on GitHub at http://github.com/wangshenguiuc/DeepSyn/ under an open source distribution license.

Background: Atrial fibrillation (AF) increases the risk of stroke 5-fold and there is rising interest to determine if AF severity or burden can further risk stratify these patients, particularly for near-term events. Using continuous remote monitoring data from cardiac implantable electronic devices, we sought to evaluate if machine learned signatures of AF burden could provide prognostic information on near-term risk of stroke when compared to conventional risk scores. Methods and Results: We retrospectively identified Veterans Health Administration serviced patients with cardiac implantable electronic device remote monitoring data and at least one day of device-registered AF. The first 30 days of remote monitoring in nonstroke controls were compared against the past 30 days of remote monitoring before stroke in cases. We trained 3 types of models on our data: (1) convolutional neural networks, (2) random forest, and (3) L1 regularized logistic regression (LASSO). We calculated the CHA 2 DS 2 -VASc score for each patient and compared its performance against machine learned indices based on AF burden in separate test cohorts. Finally, we investigated the effect of combining our AF burden models with CHA 2 DS 2 -VASc. We identified 3114 nonstroke controls and 71 stroke cases, with no significant differences in baseline characteristics. Random forest performed the best in the test data set (area under the curve [AUC]=0.662) and convolutional neural network in the validation dataset (AUC=0.702), whereas CHA 2 DS 2 -VASc had an AUC of 0.5 or less in both data sets. Combining CHA 2 DS 2 -VASc with random forest and convolutional neural network yielded a validation AUC of 0.696 and test AUC of 0.634, yielding the highest average AUC on nontraining data. Conclusions: This proof-of-concept study found that machine learning and ensemble methods that incorporate daily AF burden signature provided incremental prognostic value for risk stratification beyond CHA 2 DS 2 -VASc for near-term risk of stroke.

In 2017, drug abuse caused over 73,000 deaths in the United States. Emerging drug abuse trends are identified through community surveillance programs, medical claims data, and other healthcare system data. Social media currently exists outside that system and shows promise as an alternative means of monitoring drug abuse, but the data is massive and noisy. Initial attempts to use social media data have relied on exact text matches to drugs of interest, and therefore suffer from the gap between formal drug lexicons and the informal nature of social media. The Reddit comment archive represents an ideal corpus for bridging this gap. We trained a word embedding model, RedMed, to identify and retrieve health entities from Reddit data. We compare the performance of our consumer-generated corpus against publicly available models trained on expert-generated corpora. Our automated pipeline achieves an accuracy of 0.88 and a specificity of >0.9 when classifying across four different term classes. Of all drug mentions, an average of 79% (±0.5%) were exact matches to a generic or trademark drug name, 14% (±0.5%) were misspellings, 6.4% (±0.3%) were synonyms, and 0.13% (±0.05%) were pill marks. We find that our system captures an additional 20% of mentions; these would have been missed by approaches that rely solely on exact string matches. We provide a lexicon of misspellings and synonyms for 2,978 drugs and a word embedding model trained on a health-oriented subset of Reddit.

Objective Non-small cell lung cancer is a leading cause of cancer death worldwide, and histopathological evaluation plays the primary role in its diagnosis. However, the morphological patterns associated with the molecular subtypes have not been systematically studied. To bridge this gap, we developed a quantitative histopathology analytic framework to identify the types and gene expression subtypes of non-small cell lung cancer objectively. Materials and methods We processed whole-slide histopathology images of lung adenocarcinoma (n = 427) and lung squamous cell carcinoma patients (n = 457) in the Cancer Genome Atlas. We built convolutional neural networks to classify histopathology images, evaluated their performance by the areas under the receiver-operating characteristic curves (AUCs), and validated the results in an independent cohort (n = 125). Results To establish neural networks for quantitative image analyses, we first built convolutional neural network models to identify tumor regions from adjacent dense benign tissues (AUCs > 0.935) and recapitulated expert pathologists' diagnosis (AUCs > 0.877), with the results validated in an independent cohort (AUCs = 0.726-0.864). We further demonstrated that quantitative histopathology morphology features identified the major transcriptomic subtypes of both adenocarcinoma and squamous cell carcinoma (P Discussion Our study is the first to classify the transcriptomic subtypes of non-small cell lung cancer using fully automated machine learning methods. Our approach does not rely on prior pathology knowledge and can discover novel clinically relevant histopathology patterns objectively. The developed procedure is generalizable to other tumor types or diseases.