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

Machine Learning Techniques for Predicting Drug-Related Side Effects: A Scoping Review

BACKGROUND

Drug safety relies on advanced methods for timely and accurate prediction of side effects. To tackle this requirement, this scoping review examines machine-learning approaches for predicting drug-related side effects with a particular focus on chemical, biological, and phenotypical features.

METHODS

This was a scoping review in which a comprehensive search was conducted in various databases from 1 January 2013 to 31 December 2023.

RESULTS

The results showed the widespread use of Random Forest, k-nearest neighbor, and support vector machine algorithms. Ensemble methods, particularly random forest, emphasized the significance of integrating chemical and biological features in predicting drug-related side effects.

CONCLUSIONS

This review article emphasized the significance of considering a variety of features, datasets, and machine learning algorithms for predicting drug-related side effects. Ensemble methods and Random Forest showed the best performance and combining chemical and biological features improved prediction. The results suggested that machine learning techniques have some potential to improve drug development and trials. Future work should focus on specific feature types, selection techniques, and graph-based methods for even better prediction.

A patient safety knowledge graph supporting vaccine product development

BACKGROUND

Knowledge graphs are well-suited for modeling complex, unstructured, and multi-source data and facilitating their analysis. During the COVID-19 pandemic, adverse event data were integrated into a knowledge graph to support vaccine safety surveillance and nimbly respond to urgent health authority questions. Here, we provide details of this post-marketing safety system using public data sources. In addition to challenges with varied data representations, adverse event reporting on the COVID-19 vaccines generated an unprecedented volume of data; an order of magnitude larger than adverse events for all previous vaccines. The Patient Safety Knowledge Graph (PSKG) is a robust data store to accommodate the volume of adverse event data and harmonize primary surveillance data sources.

METHODS

We designed a semantic model to represent key safety concepts. We built an extract-transform-load (ETL) data pipeline to parse and import primary public data sources; align key elements such as vaccine names; integrated the Medical Dictionary for Regulatory Activities (MedDRA); and applied quality metrics. PSKG is deployed in a Neo4J graph database, and made available via a web interface and Application Programming Interfaces (APIs).

RESULTS

We import and align adverse event data and vaccine exposure data from 250 countries on a weekly basis, producing a graph with 4,340,980 nodes and 30,544,475 edges as of July 1, 2022. PSKG is used for ad-hoc analyses and periodic reporting for several widely available COVID-19 vaccines. Analysis code using the knowledge graph is 80% shorter than an equivalent implementation written entirely in Python, and runs over 200 times faster.

CONCLUSIONS

Organizing safety data into a concise model of nodes, properties, and edge relationships has greatly simplified analysis code by removing complex parsing and transformation algorithms from individual analyses and instead managing these centrally. The adoption of the knowledge graph transformed how the team answers key scientific and medical questions. Whereas previously an analysis would involve aggregating and transforming primary datasets from scratch to answer a specific question, the team can now iterate easily and respond as quickly as requests evolve (e.g., "Produce vaccine-X safety profile for adverse event-Y by country instead of age-range").

Machine learning prediction of side effects for drugs in clinical trials

Early and accurate detection of side effects is critical for the clinical success of drugs under development. Here, we aim to predict unknown side effects for drugs with a small number of side effects identified in randomized controlled clinical trials. Our machine learning framework, the geometric self-expressive model (GSEM), learns globally optimal self-representations for drugs and side effects from pharmacological graph networks. We show the usefulness of the GSEM on 505 therapeutically diverse drugs and 904 side effects from multiple human physiological systems. Here, we also show a data integration strategy that could be adopted to improve the ability of side effect prediction models to identify unknown side effects that might only appear after the drug enters the market.

Rituximab and Pyoderma Gangrenosum: An Investigation of Disproportionality Using a Systems Biology-Informed Approach in the FAERS Database

BACKGROUND

Studies have found an increased risk of pyoderma gangrenosum associated with rituximab. The structural properties and pharmacological action of rituximab may affect the risk of pyoderma gangrenosum. Additionally, pyoderma gangrenosum is associated with autoimmune disorders for which rituximab is indicated.

OBJECTIVE

We aimed to determine whether rituximab is disproportionally associated with pyoderma gangrenosum using a systems biology-informed approach.

METHODS

Adverse event reports were extracted from the US Food and Drug Administration Adverse Event Reporting System (FAERS, 2013-20). The Bayesian Confidence Propagation Neural Network Information Component was used to test for disproportionality. Comparators used to determine potential causal pathways included all other medicines, all medicines with a similar structure (monoclonal antibodies), all medicines with the same pharmacological target (CD20 antagonists) and all medicines used for the same indication(s) as rituximab.

RESULTS

Thirty-two pyoderma gangrenosum cases were identified, 62.5% were female, with a median age of 48 years. There was an increased association of pyoderma gangrenosum with rituximab compared with all other medicines (exponentiated Information Component 6.75, 95% confidence interval (CI) 4.66-9.23). No association was observed when the comparator was either monoclonal antibodies or CD20 antagonists. Conditions for which an association of pyoderma gangrenosum with rituximab was observed were multiple sclerosis (6.68, 95% CI 1.63-15.15), rheumatoid arthritis (2.67, 95% CI 1.14-4.80) and non-Hodgkin's lymphoma (2.94, 95% CI 1.80-3.73).

CONCLUSIONS

Pyoderma gangrenosum was reported more frequently with rituximab compared with all other medicines. The varying results when restricting medicines for the same condition suggest the potential for confounding by indication. Post-market surveillance of biologic medicines in  FAERS should consider a multi-faceted approach, particularly when the outcome of interest is associated with the underlying immune condition being treated by the medicine of interest.

Advancing drug safety science by integrating molecular knowledge with post-marketing adverse event reports

Promising drug development efforts may frequently fail due to unintended adverse reactions. Several methods have been developed to analyze such data, aiming to improve pharmacovigilance and drug safety. In this work, we provide a brief review of key directions to quantitatively analyzing adverse events and explore the potential of augmenting these methods using additional molecular data descriptors. We argue that molecular expansion of adverse event data may provide a path to improving the insights gained through more traditional pharmacovigilance approaches. Examples include the ability to assess statistical relevance with respect to underlying biomolecular mechanisms, the ability to generate plausible causative hypotheses and/or confirmation where possible, the ability to computationally study potential clinical trial designs and/or results, as well as the further provision of advanced features incorporated in innovative methods, such as machine learning. In summary, molecular data expansion provides an elegant way to extend mechanistic modeling, systems pharmacology, and patient‐centered approaches for the assessment of drug safety. We anticipate that such advances in real‐world data informatics and outcome analytics will help to better inform public health, via the improved ability to prospectively understand and predict various types of drug‐induced molecular perturbations and adverse events.

Bioinformatics Accelerates the Major Tetrad: A Real Boost for the Pharmaceutical Industry

With advanced technology and its development, bioinformatics is one of the avant-garde fields that has managed to make amazing progress in the pharmaceutical-medical field by modeling the infrastructural dimensions of healthcare and integrating computing tools in drug innovation, facilitating prevention, detection/more accurate diagnosis, and treatment of disorders, while saving time and money. By association, bioinformatics and pharmacovigilance promoted both sample analyzes and interpretation of drug side effects, also focusing on drug discovery and development (DDD), in which systems biology, a personalized approach, and drug repositioning were considered together with translational medicine. The role of bioinformatics has been highlighted in DDD, proteomics, genetics, modeling, miRNA discovery and assessment, and clinical genome sequencing. The authors have collated significant data from the most known online databases and publishers, also narrowing the diversified applications, in order to target four major areas (tetrad): DDD, anti-microbial research, genomic sequencing, and miRNA research and its significance in the management of current pandemic context. Our analysis aims to provide optimal data in the field by stratification of the information related to the published data in key sectors and to capture the attention of researchers interested in bioinformatics, a field that has succeeded in advancing the healthcare paradigm by introducing developing techniques and multiple database platforms, addressed in the manuscript.

Investigating ADR mechanisms with Explainable AI: a feasibility study with knowledge graph mining

BACKGROUND

Adverse drug reactions (ADRs) are statistically characterized within randomized clinical trials and postmarketing pharmacovigilance, but their molecular mechanism remains unknown in most cases. This is true even for hepatic or skin toxicities, which are classically monitored during drug design. Aside from clinical trials, many elements of knowledge about drug ingredients are available in open-access knowledge graphs, such as their properties, interactions, or involvements in pathways. In addition, drug classifications that label drugs as either causative or not for several ADRs, have been established.

METHODS

We propose in this paper to mine knowledge graphs for identifying biomolecular features that may enable automatically reproducing expert classifications that distinguish drugs causative or not for a given type of ADR. In an Explainable AI perspective, we explore simple classification techniques such as Decision Trees and Classification Rules because they provide human-readable models, which explain the classification itself, but may also provide elements of explanation for molecular mechanisms behind ADRs. In summary, (1) we mine a knowledge graph for features; (2) we train classifiers at distinguishing, on the basis of extracted features, drugs associated or not with two commonly monitored ADRs: drug-induced liver injuries (DILI) and severe cutaneous adverse reactions (SCAR); (3) we isolate features that are both efficient in reproducing expert classifications and interpretable by experts (i.e., Gene Ontology terms, drug targets, or pathway names); and (4) we manually evaluate in a mini-study how they may be explanatory.

RESULTS

Extracted features reproduce with a good fidelity classifications of drugs causative or not for DILI and SCAR (Accuracy = 0.74 and 0.81, respectively). Experts fully agreed that 73% and 38% of the most discriminative features are possibly explanatory for DILI and SCAR, respectively; and partially agreed (2/3) for 90% and 77% of them.

CONCLUSION

Knowledge graphs provide sufficiently diverse features to enable simple and explainable models to distinguish between drugs that are causative or not for ADRs. In addition to explaining classifications, most discriminative features appear to be good candidates for investigating ADR mechanisms further.

Artificial intelligence in oncology: From bench to clinic

In the past few years, Artificial Intelligence (AI) techniques have been applied to almost every facet of oncology, from basic research to drug development and clinical care. In the clinical arena where AI has perhaps received the most attention, AI is showing promise in enhancing and automating image-based diagnostic approaches in fields such as radiology and pathology. Robust AI applications, which retain high performance and reproducibility over multiple datasets, extend from predicting indications for drug development to improving clinical decision support using electronic health record data. In this article, we review some of these advances. We also introduce common concepts and fundamentals of AI and its various uses, along with its caveats, to provide an overview of the opportunities and challenges in the field of oncology. Leveraging AI techniques productively to provide better care throughout a patient's medical journey can fuel the predictive promise of precision medicine.

Emerging approaches to polypharmacy among older adults

Polypharmacy is a major health issue for older adults. Entangled with several geriatric syndromes, including frailty, falls and cognitive decline, research focused on polypharmacy has been challenged by heterogeneity in its definition, confounding by comorbidities and limited prospective data. In this Review, we discuss varying definitions for polypharmacy and highlight the need for a uniform definition for future studies. We critically appraise strategies for reducing medication prescriptions and implementing deprescribing as a mechanism to reduce the potential harmful effects of polypharmacy. As we look to the future, we assess the role of novel analytics and high-throughput technology, including multiomics profiling, to advance research in polypharmacy and the development of new strategies for risk stratification in the age of precision medicine.

Towards the routine use of in silico screenings for drug discovery using metabolic modelling

Currently, the development of new effective drugs for cancer therapy is not only hindered by development costs, drug efficacy, and drug safety but also by the rapid occurrence of drug resistance in cancer. Hence, new tools are needed to study the underlying mechanisms in cancer. Here, we discuss the current use of metabolic modelling approaches to identify cancer-specific metabolism and find possible new drug targets and drugs for repurposing. Furthermore, we list valuable resources that are needed for the reconstruction of cancer-specific models by integrating various available datasets with genome-scale metabolic reconstructions using model-building algorithms. We also discuss how new drug targets can be determined by using gene essentiality analysis, an in silico method to predict essential genes in a given condition such as cancer and how synthetic lethality studies could greatly benefit cancer patients by suggesting drug combinations with reduced side effects.

Drug-target-ADR Network and Possible Implications of Structural Variants in Adverse Events

Adverse drug reactions (ADRs) are of major concern in drug safety. However, due to the biological complexity of human systems, understanding the underlying mechanisms involved in development of ADRs remains a challenging task. Here, we applied network sciences to analyze a tripartite network between 1000 drugs, 1407 targets, and 6164 ADRs. It allowed us to suggest drug targets susceptible to be associated to ADRs and organs, based on the system organ class (SOC). Furthermore, a score was developed to determine the contribution of a set of proteins to ADRs. Finally, we identified proteins that might increase the susceptibility of genes to ADRs, on the basis of knowledge about genomic structural variation in genes encoding proteins targeted by drugs. Such analysis should pave the way to individualize drug therapy and precision medicine.

Prediction of Adverse Drug Reactions by Combining Biomedical Tripartite Network and Graph Representation Model

As one of the primary contributors to high clinical attrition rates of drugs, toxicity evaluation is of critical significance to new drug discovery. Unsurprisingly, a vast number of computational methods have been developed at various stages of development pipeline to evaluate potential adverse drug reactions (ADRs). Despite previous success of these methods on individual ADR or certain drug family, there are great challenges to toxicity evaluation. In this study, a novel strategy was developed to predict the drug-ADR associations by combining deep learning and the biomedical tripartite network. This heterogeneous network contains biomedical linked data of three entities, for example, drugs, targets, and ADRs. For the first time, GraRep, a deep learning method for distributed representations, is introduced to learn graph representations and identify hidden features from the tripartite network which are further used for ADR prediction. Through this approach, drug-ADR associations could possibly be discovered from a systemic perspective. The accuracy of our method is 0.95 based on internal resource validation and 0.88 based on external resource validation. Moreover, our results show the prediction accuracy using the tripartite network is better than the one with bipartite network, suggesting the model performance can be improved with further enrichment on information. According to the result of 10-fold cross validation, the deep learning model outperforms two traditional methods (topology-based measures and chemical structure-based measures). Additionally, predictive models are also constructed using other deep learning methods, and comparable results are achieved. In summary, the biomedical tripartite network-based deep learning model proposed here proves to offer a promising solution for prediction of ADRs.

Computational Advances in Drug Safety: Systematic and Mapping Review of Knowledge Engineering Based Approaches

Drug Safety (DS) is a domain with significant public health and social impact. Knowledge Engineering (KE) is the Computer Science discipline elaborating on methods and tools for developing “knowledge-intensive” systems, depending on a conceptual “knowledge” schema and some kind of “reasoning” process. The present systematic and mapping review aims to investigate KE-based approaches employed for DS and highlight the introduced added value as well as trends and possible gaps in the domain. Journal articles published between 2006 and 2017 were retrieved from PubMed/MEDLINE and Web of Science® (873 in total) and filtered based on a comprehensive set of inclusion/exclusion criteria. The 80 finally selected articles were reviewed on full-text, while the mapping process relied on a set of concrete criteria (concerning specific KE and DS core activities, special DS topics, employed data sources, reference ontologies/terminologies, and computational methods, etc.). The analysis results are publicly available as online interactive analytics graphs. The review clearly depicted increased use of KE approaches for DS. The collected data illustrate the use of KE for various DS aspects, such as Adverse Drug Event (ADE) information collection, detection, and assessment. Moreover, the quantified analysis of using KE for the respective DS core activities highlighted room for intensifying research on KE for ADE monitoring, prevention and reporting. Finally, the assessed use of the various data sources for DS special topics demonstrated extensive use of dominant data sources for DS surveillance, i.e., Spontaneous Reporting Systems, but also increasing interest in the use of emerging data sources, e.g., observational healthcare databases, biochemical/genetic databases, and social media. Various exemplar applications were identified with promising results, e.g., improvement in Adverse Drug Reaction (ADR) prediction, detection of drug interactions, and novel ADE profiles related with specific mechanisms of action, etc. Nevertheless, since the reviewed studies mostly concerned proof-of-concept implementations, more intense research is required to increase the maturity level that is necessary for KE approaches to reach routine DS practice. In conclusion, we argue that efficiently addressing DS data analytics and management challenges requires the introduction of high-throughput KE-based methods for effective knowledge discovery and management, resulting ultimately, in the establishment of a continuous learning DS system.

Retrospective Toxicological Profiling of Radium-223 Dichloride for the Treatment of Bone Metastases in Prostate Cancer Using Adverse Event Data

Background and Objective: Radium-223 dichloride (Xofigo^®) is a calcium mimetic agent approved for the treatment of castration-resistant prostate cancer patients with symptomatic bone metastases and no known visceral metastatic disease. This targeted, α-particle-emitting therapy has demonstrated significant survival benefit accompanied by a favorable safety profile. Nevertheless, recent evidence suggests that its combined use with abiraterone and prednisone/prednisolone may be associated with increased risk of death and fractures. While the precise pathophysiologic mechanisms of these events are not yet clear, collecting evidence from more clinical trials and translational studies is necessary. The aim of our present study is to assess whether accessible sources of patient outcome data can help gain additional clinical insights to radium-223 dichloride’s safety profile. Materials and Methods: We performed a retrospective analysis of cases extracted from the FDA Adverse Event Reporting System and characterized side effect occurrence by using reporting ratios. Results: A total of ~1500 prostate cancer patients treated with radium-223 dichloride was identified, and side effects reported with the use of radium-223 dichloride alone or in combination with other therapeutic agents were extracted. Our analysis demonstrates that radium-223 dichloride may often come with hematological-related reactions, and that, when administered together with other drugs, its safety profile may differ. Conclusions: While more prospective studies are needed to fully characterize the toxicological profile of radium-223 dichloride, the present work constitutes perhaps the first effort to examine its safety when administered alone and in combination with other agents based on computational evidence from public real-world post marketing data.

Home-based Anti-Tuberculosis Treatment Adverse Reactions (HATTAR) study: a protocol for a prospective observational study

INTRODUCTION

Tuberculosis (TB) continues to be an important public health problem throughout much of the world. Drug treatment is the only effective treatment method, but adverse drug events (ADEs) and adverse drug reactions (ADRs) can affect medication adherence. As the number of drug-resistant TB patients and the number of anti-TB drugs have increased, it is necessary to explore the risk factors for ADEs/ADRs to reduce their occurrence. This study aims to build a home-based anti-TB treatment cohort and to recognise the incidences, prognosis and risk factors of anti-TB drug-induced ADEs/ADRs in real-world experiences.

METHODS AND ANALYSIS

This study is a multicentre, prospective observational cohort study. The study population will consist of 3200 newly diagnosed TB patients between January 2019 and December 2020. After initiating the anti-TB treatment, all patients will be followed up until finishing treatment unless they withdraw, and we will record personal drug use and signs and/or symptoms of discomfort. Patients will receive scheduled laboratory tests in designated hospitals every 2 weeks during the first 2 months, and the residual blood sample after conducting the laboratory tests will be preserved. The ADEs/ADRs will be placed into eight categories: liver dysfunction, gastrointestinal reactions, drug allergy, arthralgia or muscle pain, nervous system disorders, haematological system disorders, renal impairment and others.

ETHICS AND DISSEMINATION

This study protocol has been approved by the ethics committees of Nanjing Medical University. All patients will give written informed consent before enrollment. The findings of the study will be published in peer-reviewed journals and will be presented at national and international conferences.

Adverse Event Circumstances and the Case of Drug Interactions

Adverse events are a common and for the most part unavoidable consequence of therapeutic intervention. Nevertheless, available tomes of such data now provide us with an invaluable opportunity to study the relationship between human phenotype and drug-induced protein perturbations within a patient system. Deciphering the molecular basis of such adverse responses is not only paramount to the development of safer drugs but also presents a unique opportunity to dissect disease systems in search of novel response biomarkers, drug targets, and efficacious combination therapies. Inspired by the potential applications of this approach, we first examined adverse event circumstances reported in FAERS and then performed a molecular level interrogation of cancer patient adverse events to investigate the prevalence of drug-drug interactions in the context of patient responses. We discuss avoidable and/or preventable cases and how molecular analytics can help optimize therapeutic use of co-medications. While up to one out of three adverse events in this dataset might be explicable by iatrogenic, patient, and product/device related factors, almost half of the patients in FAERS received multiple drugs and one in four may have experienced effects attributable to drug interactions.

Examining Socioeconomic and Computational Aspects of Vaccine Pharmacovigilance

BACKGROUND

Vaccine pharmacovigilance relates to the detection of adverse events, their assessment, understanding, and prevention, and communication of their risk to the public. These activities can be tedious and long lasting for regulatory authority scientists and may be affected by community practices and public health policies. To better understand underlying challenges, we examined vaccine adverse event reports, assessed whether data-driven techniques can provide additional insight in safety characterization, and wondered on the impact of socioeconomic parameters.

METHODS

First, we integrated VAERS content with additional sources of drug and molecular data and examined reaction and outcome occurrence by using disproportionality metrics and enrichment analysis. Second, we reviewed social and behavioral determinants that may affect vaccine pharmacovigilance aspects.

RESULTS

We describe our experience in processing more than 607000 vaccine adverse event reports and report on the challenges to integrate more than 95500 VAERS medication narratives with structured information about drugs and other therapeutics or supplements. We found that only 12.6% of events were serious, while 8.97% referred to polypharmacy cases. Exacerbation of serious clinical patient outcomes was observed in 8.88% VAERS cases in which drugs may interact with vaccinations or with each other, regardless of vaccine activity interference. Furthermore, we characterized the symptoms reported in those cases and summarized reaction occurrence among vaccine-types. Last, we examine socioeconomic parameters and cost-management features, explore adverse event reporting trends, and highlight perspectives relating to the use and development of digital services, especially in the context of personalized and collaborative health-care.

CONCLUSIONS

This work provides an informative review of VAERS, identifies challenges and limitations in the processing of vaccine adverse event data, and calls for the better understanding of the socioeconomic landscape pertaining vaccine safety concerns. We expect that adoption of computational techniques for integrated safety assessment and interpretation is key not only to pharmacovigilance practice but also to stakeholders from the entire healthcare system.

PathFX provides mechanistic insights into drug efficacy and safety for regulatory review and therapeutic development

Failure to demonstrate efficacy and safety issues are important reasons that drugs do not reach the market. An incomplete understanding of how drugs exert their effects hinders regulatory and pharmaceutical industry projections of a drug's benefits and risks. Signaling pathways mediate drug response and while many signaling molecules have been characterized for their contribution to disease or their role in drug side effects, our knowledge of these pathways is incomplete. To better understand all signaling molecules involved in drug response and the phenotype associations of these molecules, we created a novel method, PathFX, a non-commercial entity, to identify these pathways and drug-related phenotypes. We benchmarked PathFX by identifying drugs' marketed disease indications and reported a sensitivity of 41%, a 2.7-fold improvement over similar approaches. We then used PathFX to strengthen signals for drug-adverse event pairs occurring in the FDA Adverse Event Reporting System (FAERS) and also identified opportunities for drug repurposing for new diseases based on interaction paths that associated a marketed drug to that disease. By discovering molecular interaction pathways, PathFX improved our understanding of drug associations to safety and efficacy phenotypes. The algorithm may provide a new means to improve regulatory and therapeutic development decisions.

In Silico Profiling of Clinical Phenotypes for Human Targets Using Adverse Event Data

We present a novel approach for the molecular transformation and analysis of patient clinical phenotypes. Building on the fact that drugs perturb the function of targets/genes, we integrated data from 8.2 million clinical reports detailing drug-induced side effects with the molecular world of drug-target information. Using this dataset, we extracted 1.8 million associations of clinical phenotypes to 770 human drug-targets. This collection is perhaps the largest phenotypic profiling reference of human targets to-date, and unique in that it enables rapid development of testable molecular hypotheses directly from human-specific information. We also present validation results demonstrating analytical utilities of the approach, including drug safety prediction, and the design of novel combination therapies. Challenging the long-standing notion that molecular perturbation studies cannot be performed in humans, our data allows researchers to capitalize on the vast tomes of clinical information available throughout the healthcare system.

Biomarkers of adverse drug reactions

Adverse drug reactions can be caused by a wide range of therapeutics. Adverse drug reactions affect many bodily organ systems and vary widely in severity. Milder adverse drug reactions often resolve quickly following withdrawal of the casual drug or sometimes after dose reduction. Some adverse drug reactions are severe and lead to significant organ/tissue injury which can be fatal. Adverse drug reactions also represent a financial burden to both healthcare providers and the pharmaceutical industry. Thus, a number of stakeholders would benefit from development of new, robust biomarkers for the prediction, diagnosis, and prognostication of adverse drug reactions. There has been significant recent progress in identifying predictive genomic biomarkers with the potential to be used in clinical settings to reduce the burden of adverse drug reactions. These have included biomarkers that can be used to alter drug dose (for example, Thiopurine methyltransferase (TPMT) and azathioprine dose) and drug choice. The latter have in particular included human leukocyte antigen (HLA) biomarkers which identify susceptibility to immune-mediated injuries to major organs such as skin, liver, and bone marrow from a variety of drugs. This review covers both the current state of the art with regard to genomic adverse drug reaction biomarkers. We also review circulating biomarkers that have the potential to be used for both diagnosis and prognosis, and have the added advantage of providing mechanistic information. In the future, we will not be relying on single biomarkers (genomic/non-genomic), but on multiple biomarker panels, integrated through the application of different omics technologies, which will provide information on predisposition, early diagnosis, prognosis, and mechanisms. Impact statement • Genetic and circulating biomarkers present significant opportunities to personalize patient therapy to minimize the risk of adverse drug reactions. ADRs are a significant heath issue and represent a significant burden to patients, healthcare providers, and the pharmaceutical industry. • This review details the current state of the art in biomarkers of ADRs (both genetic and circulating). There is still significant variability in patient response which cannot be explained by current knowledge of genetic risk factors for ADRs; however, we discussed how specific advances in genomics have the potential to yield better and more predictive models. • Many current clinically utilized circulating biomarkers of tissue injury are valid biomarkers for a number of ADRs. However, they often give little insight into the specific cell or tissue subtype which may be affected. Emerging circulating biomarkers with potential to provide greater information on the etiology/pathophysiology of ADRs are described.

Computational models for predicting drug responses in cancer research

The computational prediction of drug responses based on the analysis of multiple types of genome-wide molecular data is vital for accomplishing the promise of precision medicine in oncology. This will benefit cancer patients by matching their tumor characteristics to the most effective therapy available. As larger and more diverse layers of patient-related data become available, further demands for new bioinformatics approaches and expertise will arise. This article reviews key strategies, resources and techniques for the prediction of drug sensitivity in cell lines and patient-derived samples. It discusses major advances and challenges associated with the different model development steps. This review highlights major trends in this area, and will assist researchers in the assessment of recent progress and in the selection of approaches to emerging applications in oncology.

Knowledge graph prediction of unknown adverse drug reactions and validation in electronic health records

Unknown adverse reactions to drugs available on the market present a significant health risk and limit accurate judgement of the cost/benefit trade-off for medications. Machine learning has the potential to predict unknown adverse reactions from current knowledge. We constructed a knowledge graph containing four types of node: drugs, protein targets, indications and adverse reactions. Using this graph, we developed a machine learning algorithm based on a simple enrichment test and first demonstrated this method performs extremely well at classifying known causes of adverse reactions (AUC 0.92). A cross validation scheme in which 10% of drug-adverse reaction edges were systematically deleted per fold showed that the method correctly predicts 68% of the deleted edges on average. Next, a subset of adverse reactions that could be reliably detected in anonymised electronic health records from South London and Maudsley NHS Foundation Trust were used to validate predictions from the model that are not currently known in public databases. High-confidence predictions were validated in electronic records significantly more frequently than random models, and outperformed standard methods (logistic regression, decision trees and support vector machines). This approach has the potential to improve patient safety by predicting adverse reactions that were not observed during randomised trials.

Inter-labeler and intra-labeler variability of condition severity classification models using active and passive learning methods

BACKGROUND AND OBJECTIVES

Labeling instances by domain experts for classification is often time consuming and expensive. To reduce such labeling efforts, we had proposed the application of active learning (AL) methods, introduced our CAESAR-ALE framework for classifying the severity of clinical conditions, and shown its significant reduction of labeling efforts. The use of any of three AL methods (one well known [SVM-Margin], and two that we introduced [Exploitation and Combination_XA]) significantly reduced (by 48% to 64%) condition labeling efforts, compared to standard passive (random instance-selection) SVM learning. Furthermore, our new AL methods achieved maximal accuracy using 12% fewer labeled cases than the SVM-Margin AL method. However, because labelers have varying levels of expertise, a major issue associated with learning methods, and AL methods in particular, is how to best to use the labeling provided by a committee of labelers. First, we wanted to know, based on the labelers' learning curves, whether using AL methods (versus standard passive learning methods) has an effect on the Intra-labeler variability (within the learning curve of each labeler) and inter-labeler variability (among the learning curves of different labelers). Then, we wanted to examine the effect of learning (either passively or actively) from the labels created by the majority consensus of a group of labelers.

METHODS

We used our CAESAR-ALE framework for classifying the severity of clinical conditions, the three AL methods and the passive learning method, as mentioned above, to induce the classifications models. We used a dataset of 516 clinical conditions and their severity labeling, represented by features aggregated from the medical records of 1.9 million patients treated at Columbia University Medical Center. We analyzed the variance of the classification performance within (intra-labeler), and especially among (inter-labeler) the classification models that were induced by using the labels provided by seven labelers. We also compared the performance of the passive and active learning models when using the consensus label.

RESULTS

The AL methods: produced, for the models induced from each labeler, smoother Intra-labeler learning curves during the training phase, compared to the models produced when using the passive learning method. The mean standard deviation of the learning curves of the three AL methods over all labelers (mean: 0.0379; range: [0.0182 to 0.0496]), was significantly lower (p=0.049) than the Intra-labeler standard deviation when using the passive learning method (mean: 0.0484; range: [0.0275-0.0724). Using the AL methods resulted in a lower mean Inter-labeler AUC standard deviation among the AUC values of the labelers' different models during the training phase, compared to the variance of the induced models' AUC values when using passive learning. The Inter-labeler AUC standard deviation, using the passive learning method (0.039), was almost twice as high as the Inter-labeler standard deviation using our two new AL methods (0.02 and 0.019, respectively). The SVM-Margin AL method resulted in an Inter-labeler standard deviation (0.029) that was higher by almost 50% than that of our two AL methods The difference in the inter-labeler standard deviation between the passive learning method and the SVM-Margin learning method was significant (p=0.042). The difference between the SVM-Margin and Exploitation method was insignificant (p=0.29), as was the difference between the Combination_XA and Exploitation methods (p=0.67). Finally, using the consensus label led to a learning curve that had a higher mean intra-labeler variance, but resulted eventually in an AUC that was at least as high as the AUC achieved using the gold standard label and that was always higher than the expected mean AUC of a randomly selected labeler, regardless of the choice of learning method (including a passive learning method). Using a paired t-test, the difference between the intra-labeler AUC standard deviation when using the consensus label, versus that value when using the other two labeling strategies, was significant only when using the passive learning method (p=0.014), but not when using any of the three AL methods.

CONCLUSIONS

The use of AL methods, (a) reduces intra-labeler variability in the performance of the induced models during the training phase, and thus reduces the risk of halting the process at a local minimum that is significantly different in performance from the rest of the learned models; and (b) reduces Inter-labeler performance variance, and thus reduces the dependence on the use of a particular labeler. In addition, the use of a consensus label, agreed upon by a rather uneven group of labelers, might be at least as good as using the gold standard labeler, who might not be available, and certainly better than randomly selecting one of the group's individual labelers. Finally, using the AL methods: when provided by the consensus label reduced the intra-labeler AUC variance during the learning phase, compared to using passive learning.

Prognosis of Clinical Outcomes with Temporal Patterns and Experiences with One Class Feature Selection

Accurate prognosis of outcome events, such as clinical procedures or disease diagnosis, is central in medicine. The emergence of longitudinal clinical data, like the Electronic Health Records (EHR), represents an opportunity to develop automated methods for predicting patient outcomes. However, these data are highly dimensional and very sparse, complicating the application of predictive modeling techniques. Further, their temporal nature is not fully exploited by current methods, and temporal abstraction was recently used which results in symbolic time intervals representation. We present Maitreya, a framework for the prediction of outcome events that leverages these symbolic time intervals. Using Maitreya, learn predictive models based on the temporal patterns in the clinical records that are prognostic markers and use these markers to train predictive models for eight clinical procedures. In order to decrease the number of patterns that are used as features, we propose the use of three one class feature selection methods. We evaluate the performance of Maitreya under several parameter settings, including the one-class feature selection, and compare our results to that of atemporal approaches. In general, we found that the use of temporal patterns outperformed the atemporal methods, when representing the number of pattern occurrences.

Towards a better cancer precision medicine: systems biology meets immunotherapy

Systems biology approaches that embrace the complexity of cancer are starting to gain traction in the development of new anticancer therapeutic strategies. In this review we describe how genomic analyses are helping improve our understanding of response to immunotherapy, a front-runner in cancer treatment. We argue that systems-level approaches are needed to help understand the concerted impact of tumor-specific and immune-specific molecular features on clinical outcomes, predict responders and unravel the complexity of tumor ecosystems. This integrated approach will propel immunotherapy into the exciting world of precision medicine.

Use of Biomedical Ontologies for Integration of Biological Knowledge for Learning and Prediction of Adverse Drug Reactions

Drug-induced toxicity is a major public health concern that leads to patient morbidity and mortality. To address this problem, the Food and Drug Administration is working on the PredicTox initiative, a pilot research program on tyrosine kinase inhibitors, to build mechanistic and predictive models for drug-induced toxicity. This program involves integrating data acquired during preclinical studies and clinical trials within pharmaceutical company development programs that they have agreed to put in the public domain and in publicly available biological, pharmacological, and chemical databases. The integration process is accommodated by biomedical ontologies, a set of standardized vocabularies that define terms and logical relationships between them in each vocabulary. We describe a few programs that have used ontologies to address biomedical questions. The PredicTox effort is leveraging the experience gathered from these early initiatives to develop an infrastructure that allows evaluation of the hypothesis that having a mechanistic understanding underlying adverse drug reactions will improve the capacity to understand drug-induced clinical adverse drug reactions.

Transcriptome profiling of NIH3T3 cell lines expressing opsin and the P23H opsin mutant identifies candidate drugs for the treatment of retinitis pigmentosa

Mammalian cells are commonly employed in screening assays to identify active compounds that could potentially affect the progression of different human diseases including retinitis pigmentosa (RP), a class of inherited diseases causing retinal degeneration with compromised vision. Using transcriptome analysis, we compared NIH3T3 cells expressing wildtype (WT) rod opsin with a retinal disease-causing single P23H mutation. Surprisingly, heterologous expression of WT opsin in NIH3T3 cells caused more than a 2-fold change in 783 out of 16,888 protein coding transcripts. The perturbed genes encoded extracellular matrix proteins, growth factors, cytoskeleton proteins, glycoproteins and metalloproteases involved in cell adhesion, morphology and migration. A different set of 347 transcripts was either up- or down-regulated when the P23H mutant opsin was expressed suggesting an altered molecular perturbation compared to WT opsin. Transcriptome perturbations elicited by drug candidates aimed at mitigating the effects of the mutant protein revealed that different drugs targeted distinct molecular pathways that resulted in a similar phenotype selected by a cell-based high-throughput screen. Thus, transcriptome profiling can provide essential information about the therapeutic potential of a candidate drug to restore normal gene expression in pathological conditions.