Drug association (DDIs) prediction is also called drug interaction prediction, which refers to the interactions between drug and drug that lead to unexpected side effects when two or more drugs are taken simultaneously or successively. Previous studies on DDIs prediction using methods such as molecular representation and network embedding were extremely complex, expensive and time-consuming, and were limited in acquiring rich neighborhood information about drug entities and their surroundings during the forecasting process. A drug linkage prediction method based on knowledge graphs and hybrid neural networks was proposed based on the deficiencies of the above methods. This method is mainly based on methods such as knowledge graphs, graph convolutional network, Convolutional-BiLSTM network and attention mechanisms to solve the limitations in acquiring rich neighborhood information about KG entities during the forecasting process. It transforms drug linkage prediction research into a link prediction problem and views drug relationships with known interactions as edges in the interaction graph. It can effectively discover interactions of unknown drugs; meanwhile, performance comparisons are performed with existing DDIs prediction methods. The results show that higher performance is achieved in terms of indicators such as ACC and F1 values, which validate the effectiveness of the model. Finally, future directions in this field are proposed based on an analysis and summary of challenges faced by current DDIs predictions.

Motivation. Immunoglobulin proteins (IGP) (also called antibodies) are glycoproteins that act as B-cell receptors against external or internal antigens like viruses and bacteria. IGPs play a significant role in diverse cellular processes ranging from adhesion to cell recognition. IGP identifications via the in-silico approach are faster and more cost-effective than wet-lab technological methods. Methods. In this study, we developed an intelligent theoretical deep learning framework, IGPred-HDnet for the discrimination of IGPs and non-IGPs. Three types of promising descriptors are feature extraction based on graphical and statistical features (FEGS), amphiphilic pseudo-amino acid composition (Amp-PseAAC), and dipeptide composition (DPC) to extract the graphical, physicochemical, and sequential features. Next, the extracted attributes are evaluated through machine learning, i.e., decision tree (DT), support vector machine (SVM), k-nearest neighbour (KNN), and hierarchical deep network (HDnet) classifiers. The proposed predictor IGPred-HDnet was trained and tested using a 10-fold cross-validation and independent test. Results and Conclusion. The success rates in terms of accuracy (ACC) and Matthews correlation coefficient (MCC) of IGPred-HDnet on training and independent dataset (D_train D_test) are ACC=98.00, 99.10, and MCC=0.958, and 0.980 points, respectively. The empirical outcomes demonstrate that the IGPred-HDnet model efficacy on both datasets using the novel FEGS feature and HDnet algorithm achieved superior predictions to other existing computational models. We hope this research will provide great insights into the large-scale identification of IGPs and pharmaceutical companies in new drug design.

Drug side effect is an important entity in the biomedical field, and identifying the association of the drug-side effects is a very important issue in pharmacological studies and drug risk-benefit. Traditional side effect discovery methods are mainly based on pharmacological experiments. These methods can detect the side effects of some drugs, but the identification process is time-consuming, expensive, and fails to identify some rare side effects. In recent years, with the expansion of massive biomedical data, computational-based methods are widely developed and applied for the task of drug-side effect association(DSA) prediction. However, existing methods cannot fully utilize public biomedical databases, and the complex semantic associations between drugs and side effects are not effectively captured, which leads to suboptimal model prediction performance. In this study, we develop a novel meta-path-based graph neural network model for drug-side effect association prediction. In the proposed model, we first construct a heterogeneous information network(HIN) by fusing multiple biological datasets. And then, a novel meta-path-based feature learning module is designed to learn high-quality representations of drugs and side effects. Finally, with the learned features, the prediction module utilizes a fully connected neural network to make prediction. In addition, comprehensive experiments is conducted, the results demonstrate the effectiveness of our model, indicating that the method will be a viable approach for DSA prediction tasks.

Drug association (DDIs) prediction is also called drug interaction prediction, which refers to the interactions between drug and drug that lead to unexpected side effects when two or more drugs are taken simultaneously or successively. Previous studies on DDIs prediction using methods such as molecular representation and network embedding were extremely complex, expensive and time-consuming, and were limited in acquiring rich neighborhood information about drug entities and their surroundings during the forecasting process. A drug linkage prediction method based on knowledge graphs and hybrid neural networks was proposed based on the deficiencies of the above methods. This method is mainly based on methods such as knowledge graphs, graph convolutional network, Convolutional-BiLSTM network and attention mechanisms to solve the limitations in acquiring rich neighborhood information about KG entities during the forecasting process. It transforms drug linkage prediction research into a link prediction problem and views drug relationships with known interactions as edges in the interaction graph. It can effectively discover interactions of unknown drugs; meanwhile, performance comparisons are performed with existing DDIs prediction methods. The results show that higher performance is achieved in terms of indicators such as ACC and F1 values, which validate the effectiveness of the model. Finally, future directions in this field are proposed based on an analysis and summary of challenges faced by current DDIs predictions.

An adverse event is any abnormal clinical finding associated with the use of a therapy. Adverse events are classified by reporting an event's seriousness, expectedness, and relatedness. Monitoring patient safety is of utmost importance as more and more data becomes available. In reality, very low numbers of adverse events are reported via the official path. Chart review, voluntary reporting, computerized surveillance, and direct observation can detect adverse drug events. Medication errors are commonly seen in hospitals and need provider and system-based interventions to prevent them. The need of the hour in India is to develop and implement medication safety best practices to avoid adverse events. The utility of artificial intelligence techniques in adverse event detection remains unexplored, and their accuracy and precision need to be studied in a controlled setting. There is a need to develop predictive models to assess the likelihood of adverse reactions while testing novel pharmaceutical drugs.

BACKGROUND In the last decade, many studies have investigated on predicting the potential side effects of drugs. In general, this study could be divided into two categories; detection and prediction. Detection aims to detect the side effects of existing drugs. In Prediction, the side effects of new drugs are studied. In general, when a new drug enters the drug discovery cycle, its side effects are unknown. OBJECTIVE Despite the positive effects of drugs for the prevention, diagnosis, and treatment of diseases, the side effects of some drugs could not be ignored. Many diseases and death are reported annually due to drug side effects. In general, predicting drug side effects using laboratory methods is very costly and time-consuming. On the other hand, these methods are not able to diagnose all the drug side effects due to many limitations. METHODS We attempt to obtain effective embedding for drugs and their probable side effects using the Graph attention network. Furthermore, drug side effects links are predicted using these embedding. RESULTS Using semantic and structural properties of drug network has properly increased the prediction and further detection of side effects. The results indicate Area Under Precision-Recall of 0.7258 and 0.7184 on this research dataset, respectively. CONCLUSIONS Using positive and negative examples in loss function and self-attention deployment in Graph attention network result in more reliable embedding and better accuracy results in prediction. Moreover, the significant effect of embedding in predicting links has been shown in this study.

Predicting side effects of drug–drug interactions (DDIs) is an important task in pharmacology. The state-of-the-art methods for DDI prediction use hypergraph neural networks to learn latent representations of drugs and side effects to express high-order relationships among two interacting drugs and a side effect. The idea of these methods is that each side effect is caused by a unique combination of latent features of the corresponding interacting drugs. However, in reality, a side effect might have multiple, different mechanisms that cannot be represented by a single combination of latent features of drugs. Moreover, DDI data are sparse, suggesting that using a sparsity regularization would help to learn better latent representations to improve prediction performances. We propose SPARSE, which encodes the DDI hypergraph and drug features to latent spaces to learn multiple types of combinations of latent features of drugs and side effects, controlling the model sparsity by a sparse prior. Our extensive experiments using both synthetic and three real-world DDI datasets showed the clear predictive performance advantage of SPARSE over cutting-edge competing methods. Also, latent feature analysis over unknown top predictions by SPARSE demonstrated the interpretability advantage contributed by the model sparsity. Code and data can be accessed at https://github.com/anhnda/SPARSE. Supplementary data are available at Bioinformatics online.

The prediction of adverse drug reactions (ADR) is an important step of drug discovery and design process. Different drug properties have been employed for ADR prediction but the prediction capability of drug properties and drug functions in integrated manner is yet to be explored. In the present work, a multi-label deep neural network and MLSMOTE based methodology has been proposed for ADR prediction. The proposed methodology has been applied on SMILES Strings data of drugs, 17 molecular descriptors data of drugs and drug functions data individually and in integrated manner for ADR prediction. The experimental results shows that the SMILES Strings + drug functions has outperformed other types of data with regards to ADR prediction capability.

Recently, social implementations of artificial intelligence (AI) have been rapidly advancing. Many papers have investigated the use of AI in the field of healthcare. However, there have been few studies on the adaptation of AI to clinical pharmaceutical services. We reported attempts to adapt clinical pharmaceutical services with AI in the following areas of machine learning application in prescription audits: solutions for pharmaceutical problems via speech recognition and automatic assignment of standard code to drug name information by natural language processing. Though both were exploratory attempts, we showed the usefulness of adapting AI to clinical pharmaceutical services. AI is expected to support and alter all industries in the future, including healthcare and clinical pharmaceutical services. However, AI is not magic that can solve any problem. When using an AI-adapted program, it is necessary to be aware of its features and limitations. For the coming AI era, clinical pharmacists need to improve their AI literacy.

Social forums offer a lot of new channels for collecting patients’ opinions to construct predictive models of adverse drug reactions (ADRs) for post-marketing surveillance. However, due to the characteristics of social posts, there are many challenges still to be solved when deriving such models, mainly including problems caused by data sparseness, data features with a high-dimensionality, and term diversity in data. To tackle these crucial issues related to identifying ADRs from social posts, we perform data analytics from the perspectives of data balance, feature selection, and feature learning. Meanwhile, we design a comprehensive experimental analysis to investigate the performance of different data processing techniques and data modeling methods. Most importantly, we present a deep learning-based approach that adopts the BERT (Bidirectional Encoder Representations from Transformers) model with a new batch-wise adaptive strategy to enhance the predictive performance. A series of experiments have been conducted to evaluate the machine learning methods with both manual and automated feature engineering processes. The results prove that with their own advantages both types of methods are effective in ADR prediction. In contrast to the traditional machine learning methods, our feature learning approach can automatically achieve the required task to save the manual effort for the large number of experiments.

Identification of drug-target interactions (DTIs) plays a crucial role in drug development. Traditional laboratory-based DTI discovery is generally costly and time-consuming. Therefore, computational approaches have been developed to predict interactions between drug candidates and disease-causing proteins. We designed a novel method, termed heterogeneous information integration for DTI prediction (HIDTI), based on the concept of predicting vectors for all of unknown/unavailable heterogeneous drug- and protein-related information. We applied a residual network in HIDTI to extract features of such heterogeneous information for predicting DTIs, and tested the model using drug-based ten-fold cross-validation to examine the prediction performance for unseen drugs. As a result, HIDTI outperformed existing models using heterogeneous information, and was demonstrating that our method predicted heterogeneous information on unseen data better than other models. In conclusion, our study suggests that HIDTI has the potential to advance the field of drug development by accurately predicting the targets of new drugs.

Identification of adverse drug events (ADEs) is crucial to reduce human health risks and improve drug safety assessment. With an increasing number of biological and medical data, computational methods such as network-based methods were proposed for ADE prediction with high efficiency and low cost. However, previous network-based methods rely on the topological information of known drug-ADE networks, and hence cannot make predictions for novel compounds without any known ADE. In this study, we introduced chemical substructures to bridge the gap between the drug-ADE network and novel compounds, and developed a novel network-based method named ADENet, which can predict potential ADEs for not only drugs within the drug-ADE network, but also novel compounds outside the network. To show the performance of ADENet, we collected drug-ADE associations from a comprehensive database named MetaADEDB and constructed a series of network-based prediction models. These models obtained high area under the receiver operating characteristic curve values ranging from 0.871 to 0.947 in 10-fold cross-validation. The best model further showed high performance in external validation, which outperformed a previous network-based and a recent deep learning-based method. Using several approved drugs as case studies, we found that 32–54% of the predicted ADEs can be validated by the literature, indicating the practical value of ADENet. Moreover, ADENet is freely available at our web server named NetInfer (http://lmmd.ecust.edu.cn/netinfer). In summary, our method would provide a promising tool for ADE prediction and drug safety assessment in drug discovery and development.

Anatomical Therapeutic Chemical (ATC) classes prediction is one of the prominent activities in the costly and tedious pipeline of drug discovery where machine learning plays an important role by minimizing the cost and time of prediction. Most of the existing research have been done to predict ATC classes from the chemical-chemical association, side-effects, target proteins, gene expressions, chemical structures, drug targets, and textual information of drugs. However, the capability of 17 molecules’ properties have not yet been explored to predict drug ATC classes. The current work proposes a methodology for predicting the drug ATC classes using the 17 molecules’ properties. ATC classes prediction is a multi-label classification task and therefore, a binary relevance strategy has been employed to solve this issue with four basic machine learning classifiers, namely K-Nearest Neighbour (KNN), Extra Tree Classifier (ETC), Random Forest (RF), and Decision Tree (DT). The common problem of multi-label datasets is class imbalance which is addressed using the MLSMOTE (Multi-Label Synthetic Minority Over-Sampling Technique). The proposed methodology exhibits promising results, and it achieved the accuracy ranging from 96.90% to 98.06%, which indicates that 17 molecules’ properties are good enough in efficient prediction of ATC classes.

Background: Pharmacovigilance and safety reporting, which involve processes for monitoring the use of medicines in clinical trials, play a critical role in the identification of previously unrecognized adverse events or changes in the patterns of adverse events. Objective: This study aims to demonstrate the feasibility of automating the coding of adverse events described in the narrative section of the serious adverse event report forms to enable statistical analysis of the aforementioned patterns. Methods: We used the Unified Medical Language System (UMLS) as the coding scheme, which integrates 217 source vocabularies, thus enabling coding against other relevant terminologies such as the International Classification of Diseases–10th Revision, Medical Dictionary for Regulatory Activities, and Systematized Nomenclature of Medicine). We used MetaMap, a highly configurable dictionary lookup software, to identify the mentions of the UMLS concepts. We trained a binary classifier using Bidirectional Encoder Representations from Transformers (BERT), a transformer-based language model that captures contextual relationships, to differentiate between mentions of the UMLS concepts that represented adverse events and those that did not. Results: The model achieved a high F1 score of 0.8080, despite the class imbalance. This is 10.15 percent points lower than human-like performance but also 17.45 percent points higher than that of the baseline approach. Conclusions: These results confirmed that automated coding of adverse events described in the narrative section of serious adverse event reports is feasible. Once coded, adverse events can be statistically analyzed so that any correlations with the trialed medicines can be estimated in a timely fashion.

The COVID-19 pandemic has wreaked havoc in every country in the world, with serious health-related, economic, and social consequences. Since its outbreak in March 2020, many researchers from different fields have joined forces to provide a wide range of solutions, and the support for this work from artificial intelligence (AI) and other emerging concepts linked to intelligent data analysis has been decisive. The enormous amount of research and the high number of publications during this period makes it difficult to obtain an overall view of the different applications of AI to the management of COVID-19 and an understanding of how research in this field has been evolving. Therefore, in this paper, we carry out a scientometric analysis of this area supported by text mining, including a review of 18,955 publications related to AI and COVID-19 from the Scopus database from March 2020 to June 2021 inclusive. For this purpose, we used VOSviewer software, which was developed by researchers at Leiden University in the Netherlands. This allowed us to examine the exponential growth in research on this issue and its distribution by country, and to highlight the clear hegemony of the United States (USA) and China in this respect. We used an automatic process to extract topics of research interest and observed that the most important current lines of research focused on patient-based solutions. We also identified the most relevant journals in terms of the COVID-19 pandemic, demonstrated the growing value of open-access publication, and highlighted the most influential authors by means of an analysis of citations and co-citations. This study provides an overview of the current status of research on the application of AI to the pandemic.

Since the emergence of deep learning-based chatbots for knowledge services, numerous research and development projects have been conducted in various industries. A high demand for chatbots has drastically increased the global market size; however, the limited functional scalability of open-domain chatbots is a challenge to their application to industries. Moreover, as most chatbot frameworks employ English, it is necessary to create chatbots customized for other languages. To address this problem, this paper proposes KoRASA as a pipeline-optimization method, which uses a deep learning-based open-source chatbot framework to understand the Korean language. KoRASA is a closed-domain chatbot that is applicable across a wide range of industries in Korea. KoRASAs operation consists of four stages: tokenization, featurization, intent classification, and entity extraction. The accuracy and F1-score of KoRASA were measured based on datasets taken from common tasks carried out in most industrial fields. The algorithm for intent classification and entity extraction was optimized. The accuracy and F1-score were 98.2 and 98.4 for intent classification and 97.4 and 94.7 for entity extraction, respectively. Furthermore, these results are better than those achieved by existing models. Accordingly, KoRASA can be applied to various industries, including mobile services based on closed-domain chatbots using Korean, robotic process automation (RPA), edge computing, and Internet of Energy (IoE) services.

Drug safety research asks causal questions but relies on observational data. Confounding bias threatens the reliability of studies using such data. The successful control of confounding requires knowledge of variables called confounders affecting both the exposure and outcome of interest. Causal knowledge of dynamic biological systems is complex and challenging. Fortunately, computable knowledge mined from the literature may hold clues about confounders. In this paper, we tested the hypothesis that incorporating literature-derived confounders can improve causal inference from observational data. We introduce two methods (semantic vector-based and string-based confounder search) that query literature-derived information for confounder candidates to control, using SemMedDB, a database of computable knowledge mined from the biomedical literature. These methods search SemMedDB for confounders by applying semantic constraint search for indications treated by the drug (exposure), that are also known to cause the adverse event (outcome). We then include the literature-derived confounder candidates in statistical and causal models derived from free-text clinical notes. For evaluation, we use a reference dataset widely used in drug safety containing labeled pairwise relationships between drugs and adverse events and attempt to rediscover these relationships from a corpus of 2.2M NLP-processed free-text clinical notes. We employ standard adjustment and causal inference procedures to predict and estimate causal effects by informing the models with varying numbers of literature-derived confounders and instantiating the exposure, outcome, and confounder variables in the models with dichotomous EHR-derived data. Finally, we compare the results from applying these procedures with naive measures of association (χ2 and reporting odds ratio) and with each other. We found semantic vector-based search to be superior to string-based search at reducing confounding bias. However, the effect of including more rather than fewer literature-derived confounders was inconclusive. We recommend using targeted learning estimation methods that can address treatment-confounder feedback, where confounders that also behave as intermediate variables, and engaging subject-matter experts to adjudicate the handling of problematic confounders.

Pharmacovigilance, the scientific discipline pertaining to drug safety, has been studied extensively and is progressing continuously. In this field, medical informatics techniques and interpretation play important roles, and appropriate approaches are required. In this study, we investigated and analyzed the trends of pharmacovigilance systems, especially the data collection, detection, assessment, and monitoring processes. We used PubMed to collect papers on pharmacovigilance published over the past 10 years, and analyzed a total of 40 significant papers to determine the characteristics of the databases and data analysis methods used to identify drug safety indicators. Through systematic reviews, we identified the difficulty of standardizing data and terminology and establishing an adverse drug reactions (ADR) evaluation system in pharmacovigilance, and their corresponding implications. We found that appropriate methods and guidelines for active pharmacovigilance using medical big data are still required and should continue to be developed.

Background: Preventing adverse drug reactions (ADRs) is imperative for the safety of the people. The problem of under-reporting the ADRs has been prevalent across the world, making it difficult to develop the prediction models, which are unbiased. As a result, most of the models are skewed to the negative samples leading to high accuracy but poor performance in other metrics such as precision, recall, F1 score, and AUROC score. Objective: In this work, we have proposed a novel way of predicting the ADRs by balancing the dataset. Method: The whole data set has been partitioned into balanced smaller data sets. SVMs with optimal kernel have been learned using each of the balanced data sets and the prediction of given ADR for the given drug has been obtained by voting from the ensembled optimal SVMs learned. Results: We have found that results are encouraging and comparable with the competing methods in the literature and obtained the average sensitivity of 0.97 for all the ADRs. The model has been interpreted and explained with SHAP values by various plots. Conclusion: A novel way of predicting ADRs by balancing the dataset has been proposed thereby reducing the effect of unbalanced datasets.

Social networks become widely used for understanding patients shared experiences, and reaching a vast audience in a matter of seconds. In particular, many health-related organizations used sentiment analysis to automatically reporting treatment issues, drug misuse, new infectious disease symptoms. Few approaches have proposed in this matter, especially for detecting different drug reaction descriptions from patients generated narratives on social networks. Most of them consisted of only detecting adverse drug reaction(ADR), but may fail to retrieve other aspect, e.g, the beneficial drug reaction or drug retroviral effects such as “relieve intraocular pressure associated with glaucoma”. In this study, we propose to develop an encoder-decoder for drug reaction discrimination that involves an enhanced distributed biomedical representation from controlled medical vocabulary such as PubMed and Clinical note MIMIC III. The embedding mechanism primarily leverages contextual information and learn from predefined clinical relationships in term of medical conditions in order to define possible drug reaction of individual meaning and multi-word expressions in the field of distributional semantics configuration that clarifies sentence's similarity in the same contextual target space, which are further share semantically common drug description meanings. Furthermore, the bidirectional sentiment inductive model are created to enhance drug reactions vectorization from real-world patients description whereby achieved higher performance in terms of disambiguating false positive and/or negative assessments. As a result, we achieved an 85.2% accuracy performance and the architecture shows a well-encoding of real-world drug entities descriptions.

Summary MetaADEDB is an online database we developed to integrate comprehensive information on adverse drug events (ADEs). The first version of MetaADEDB was released in 2013 and has been widely used by researchers. However, it has not been updated for more than seven years. Here, we reported its second version by collecting more and newer data from the U.S. FDA Adverse Event Reporting System (FAERS) and Canada Vigilance Adverse Reaction Online Database, in addition to the original three sources. The new version consists of 744 709 drug–ADE associations between 8498 drugs and 13 193 ADEs, which has an over 40% increase in drug–ADE associations compared to the previous version. Meanwhile, we developed a new and user-friendly web interface for data search and analysis. We hope that MetaADEDB 2.0 could provide a useful tool for drug safety assessment and related studies in drug discovery and development. Availability and implementation The database is freely available at: http://lmmd.ecust.edu.cn/metaadedb/. Supplementary information Supplementary data are available at Bioinformatics online.

Deep learning (DL) algorithms are a subset of machine learning algorithms with the aim of modeling complex mapping between a set of elements and their classes. In parallel to the advance in revealing the molecular bases of diseases, a notable innovation has been undertaken to apply DL in data/libraries management, reaction optimizations, differentiating uncertainties, molecule constructions, creating metrics from qualitative results, and prediction of structures or interactions. From source identification to lead discovery and medicinal chemistry of the drug candidate, drug delivery, and modification, the challenges can be subjected to artificial intelligence algorithms to aid in the generation and interpretation of data. Discovery and design approach, both demand automation, large data management and data fusion by the advance in high-throughput mode. The application of DL can accelerate the exploration of drug mechanisms, finding novel indications for existing drugs (drug repositioning), drug development, and preclinical and clinical studies. The impact of DL in the workflow of drug discovery, design, and their complementary tools are highlighted in this review. Additionally, the type of DL algorithms used for this purpose, and their pros and cons along with the dominant directions of future research are presented.

Background Artificial intelligence (AI) and the Internet of Intelligent Things (IIoT) are promising technologies to prevent the concerningly rapid spread of coronavirus disease (COVID-19) and to maximize safety during the pandemic. With the exponential increase in the number of COVID-19 patients, it is highly possible that physicians and health care workers will not be able to treat all cases. Thus, computer scientists can contribute to the fight against COVID-19 by introducing more intelligent solutions to achieve rapid control of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the virus that causes the disease. Objective The objectives of this review were to analyze the current literature, discuss the applicability of reported ideas for using AI to prevent and control COVID-19, and build a comprehensive view of how current systems may be useful in particular areas. This may be of great help to many health care administrators, computer scientists, and policy makers worldwide. Methods We conducted an electronic search of articles in the MEDLINE, Google Scholar, Embase, and Web of Knowledge databases to formulate a comprehensive review that summarizes different categories of the most recently reported AI-based approaches to prevent and control the spread of COVID-19. Results Our search identified the 10 most recent AI approaches that were suggested to provide the best solutions for maximizing safety and preventing the spread of COVID-19. These approaches included detection of suspected cases, large-scale screening, monitoring, interactions with experimental therapies, pneumonia screening, use of the IIoT for data and information gathering and integration, resource allocation, predictions, modeling and simulation, and robotics for medical quarantine. Conclusions We found few or almost no studies regarding the use of AI to examine COVID-19 interactions with experimental therapies, the use of AI for resource allocation to COVID-19 patients, or the use of AI and the IIoT for COVID-19 data and information gathering/integration. Moreover, the adoption of other approaches, including use of AI for COVID-19 prediction, use of AI for COVID-19 modeling and simulation, and use of AI robotics for medical quarantine, should be further emphasized by researchers because these important approaches lack sufficient numbers of studies. Therefore, we recommend that computer scientists focus on these approaches, which are still not being adequately addressed.

Background Licensed drugs may cause unexpected adverse reactions in patients, resulting in morbidity, risk of mortality, therapy disruptions, and prolonged hospital stays. Officially approved drug package inserts list the adverse reactions identified from randomized controlled clinical trials with high evidence levels and worldwide postmarketing surveillance. Formal representation of the adverse drug reaction (ADR) enclosed in semistructured package inserts will enable deep recognition of side effects and rational drug use, substantially reduce morbidity, and decrease societal costs. Objective This paper aims to present an ontological organization of traceable ADR information extracted from licensed package inserts. In addition, it will provide machine-understandable knowledge for bioinformatics analysis, semantic retrieval, and intelligent clinical applications. Methods Based on the essential content of package inserts, a generic ADR ontology model is proposed from two dimensions (and nine subdimensions), covering the ADR information and medication instructions. This is followed by a customized natural language processing method programmed with Python to retrieve the relevant information enclosed in package inserts. After the biocuration and identification of retrieved data from the package insert, an ADR ontology is automatically built for further bioinformatic analysis. Results We collected 165 package inserts of quinolone drugs from the National Medical Products Administration and other drug databases in China, and built a specialized ADR ontology containing 2879 classes and 15,711 semantic relations. For each quinolone drug, the reported ADR information and medication instructions have been logically represented and formally organized in an ADR ontology. To demonstrate its usage, the source data were further bioinformatically analyzed. For example, the number of drug-ADR triples and major ADRs associated with each active ingredient were recorded. The 10 ADRs most frequently observed among quinolones were identified and categorized based on the 18 categories defined in the proposal. The occurrence frequency, severity, and ADR mitigation method explicitly stated in package inserts were also analyzed, as well as the top 5 specific populations with contraindications for quinolone drugs. Conclusions Ontological representation and organization using officially approved information from drug package inserts enables the identification and bioinformatic analysis of adverse reactions caused by a specific drug with regard to predefined ADR ontology classes and semantic relations. The resulting ontology-based ADR knowledge source classifies drug-specific adverse reactions, and supports a better understanding of ADRs and safer prescription of medications.

Introduction: Confounding bias threatens the reliability of observational studies and poses a significant scientific challenge. This paper introduces a framework for identifying confounding factors by exploiting literature-derived computable knowledge. In previous work, we have shown that semantic constraint search over computable knowledge extracted from the literature can be useful for reducing confounding bias in statistical models of EHR-derived observational clinical data. We hypothesize that adjustment sets of literature-derived confounders could also improve causal inference.Methods: We introduce two methods (semantic vectors and string-based confounder search) that query the literature for potential confounders and use this information to build models from EHR-derived data to more accurately estimate causal effects. These methods search SemMedDB for indications TREATED BY the drug that is also known to CAUSE the adverse event. For evaluation, we attempt to rediscover associations in a publicly available reference dataset containing expected pairwise relationships between drugs and adverse events from empirical data derived from a corpus of 2.2M EHR-derived clinical notes. For our knowledge-base, we use SemMedDB, a database of computable knowledge mined from the biomedical literature. Using standard adjustment and causal inference procedures on dichotomous drug exposures, confounders, and adverse event outcomes, varying numbers of literature-derived confounders are combined with EHR data to predict and estimate causal effects in light of the literature-derived confounders. We then compare the performance of the new methods with naive (χ², reporting odds ratio) measures of association.Results and Conclusions: Logistic regression with ten vector space-derived confounders achieved the most improvement with AUROC of 0.628 (95% CI: [0.556,0.720]), compared with baseline χ²0.507 (95% CI: [0.431,0.583]). Bias reduction was improved more often in modeling methods using more rather than less information, and using semantic vector rather than string-based search. We found computable knowledge useful for improving automated causal inference, and identified opportunities for further improvement, including a role for adjudicating literature-derived confounders by subject matter experts.Graphical Highlights: Access to causal background knowledge is required for causal learning to scale to large datasets. We introduce a framework for identifying confounders to enhance causal inference from EHR. We search computable knowledge for indications TREATED BY the drug that CAUSE the outcome. Literature-derived confounders reduce confounding bias in EHR data. Structured knowledge helps interpret and explain data captured in clinical narratives.

Understanding the process of drug repurposing is critically significant for drug development. In this paper, we employ extracted bio-entities to detect the features of different phases in drug repurposing. We proposed a transparent and easy entitymetric indicator for bio-entities, i.e., Popularity Index, to quantify and visualize the dynamic changes in academic interests of bio-entities. By taking aspirin as an example, the results display specific profiles of drug repurposing and the evolution of bio-entities in the different phases of drug research, which would potentially be valuable for pharmaceutical companies and scholars to successfully discover and repurpose drugs.

MotivationDrug-drug interactions (DDIs) are complex processes which may depend on many clinical and non-clinical factors. Identifying and distinguishing ways in which drugs interact remains a challenge. To minimize DDIs and to personalize treatment based on accurate stratification of patients, it is crucial that mechanisms of interaction can be identified. Most DDIs are a consequence of metabolic mechanisms of interaction, but DDIs with different mechanisms occur less frequently and are therefore more difficult to identify.ResultsWe developed a method (D4) for computationally identifying potential DDIs and determining whether they interact based on one of eleven mechanisms of interaction. D4 predicts DDIs and their mechanisms through features that are generated through a deep learning approach from phenotypic and functional knowledge about drugs, their side-effects and targets. Our findings indicate that our method is able to identify known DDIs with high accuracy and that D4 can determine mechanisms of interaction. We also identify numerous novel and potential DDIs for each mechanism of interaction and evaluate our predictions using DDIs from adverse event reporting systems.Availabilityhttps://github.com/bio-ontology-research-group/[email protected] and [email protected]

Motivation: Adverse drug reaction (ADR) or drug side effect studies play a crucial role in drug discovery. Recently, with the rapid increase of both clinical and non-clinical data, machine learning methods have emerged as prominent tools to support analyzing and predicting ADRs. Nonetheless, there are still remaining challenges in ADR studies.Results: In this paper, we summarized ADR data sources and review ADR studies in three tasks: drug-ADR benchmark data creation, drug–ADR prediction and ADR mechanism analysis. We focused on machine learning methods used in each task and then compare performances of the methods on the drug–ADR prediction task. Finally, we discussed open problems for further ADR studies.Availability: Data and code are available at https://github.com/anhnda/ADRPModels.

Drug failures due to unforeseen adverse effects at clinical trials pose health risks for the participants and lead to substantial financial losses. Side effect prediction algorithms have the potential to guide the drug design process. LINCS L1000 dataset provides a vast resource of cell line gene expression data perturbed by different drugs and creates a knowledge base for context specific features. The state-of-the-art approach that aims at using context specific information relies on only the high-quality experiments in LINCS L1000 and discards a large portion of the experiments. In this study, our goal is to boost the prediction performance by utilizing this data to its full extent. We experiment with 5 deep learning architectures. We find that a multi-modal architecture produces the best predictive performance among multi-layer perceptron-based architectures when drug chemical structure (CS), and the full set of drug perturbed gene expression profiles (GEX) are used as modalities. Overall, we observe that the CS is more informative than the GEX. A convolutional neural network-based model that uses only SMILES string representation of the drugs achieves the best results and provides 13.0% macro-AUC and 3.1% micro-AUC improvements over the state-of-the-art. We also show that the model is able to predict side effect-drug pairs that are reported in the literature but was missing in the ground truth side effect dataset. DeepSide is available at http://github.com/OnurUner/DeepSide.

Drug-Drug Interaction (DDI) prediction is one of the most critical issues in drug development and health. Proposing appropriate computational methods for predicting unknown DDI with high precision is challenging. We proposed "NDD: Neural network-based method for drug-drug interaction prediction" for predicting unknown DDIs using various information about drugs. Multiple drug similarities based on drug substructure, target, side effect, off-label side effect, pathway, transporter, and indication data are calculated. At first, NDD uses a heuristic similarity selection process and then integrates the selected similarities with a nonlinear similarity fusion method to achieve high-level features. Afterward, it uses a neural network for interaction prediction. The similarity selection and similarity integration parts of NDD have been proposed in previous studies of other problems. Our novelty is to combine these parts with new neural network architecture and apply these approaches in the context of DDI prediction. We compared NDD with six machine learning classifiers and six state-of-the-art graph-based methods on three benchmark datasets. NDD achieved superior performance in cross-validation with AUPR ranging from 0.830 to 0.947, AUC from 0.954 to 0.994 and F-measure from 0.772 to 0.902. Moreover, cumulative evidence in case studies on numerous drug pairs, further confirm the ability of NDD to predict unknown DDIs. The evaluations corroborate that NDD is an efficient method for predicting unknown DDIs. The data and implementation of NDD are available at https://github.com/nrohani/NDD.

Drug-drug interactions (DDIs) and their associated adverse drug reactions (ADRs) represent a significant detriment to the public health. Existing research on DDIs is primarily focused on pairwise DDI detection and prediction. It is highly needed to develop effective computational tools for high-order DDI prediction. Here we show that deep learning can be effectively utilized to predict ADRs induced from high-order DDIs. In this manuscript, we present a deep learning model D3I for cardinality-invariant and order-invariant high-order DDI prediction. The D3I models achieve 0.740 F1 value and 0.847 AUC value on balanced high-order DDI prediction, and outperform other models on order-2 DDI prediction. These results demonstrate the strong potential of D3I and deep learning models in tackling the prediction problems of high-order DDIs and their induced ADRs. In addition, D3I is able to derive single drug representations, which conform to our current knowledge on single drugs, from their behaviors in drug combinations. D3I can also correctly predict ADRs for drug combinations in which no single drugs on their own induce ADRs, and improve ADR prediction on drug pairs by learning from all drug combinations. To the best of our knowledge, D3I is the first deep model for high-order DDI prediction.

Adverse drug events (ADEs) detection is the critical concern in the field of pharmacovigilance, and it is also necessary to optimize the ADEs prediction to reduce the drug related morbidity and mortality. Here we propose a novel methods of data mining algorithms directed predictive pharmacosafety networks (PPNs) to compare their predictive performance and investigate the differences between data mining algorithms. The combinations of 152 cancer drugs and 633 ADEs in the 2010 FDA Adverse Event Reporting System(FAERS) data is the training data, and 2011-2015 FAERS data is the validation data. We find that performance of empirical Bayes geometric mean (EBGM) is closer to proportional reporting ratio (PRR), and greater than reporting odds ratio (ROR) in ADE detection. Further, only information component (IC) directed the PPNs have better predictive performance comparing to other data mining algorithms, the predictive performance of which reaches to AUROC =0.908 comparing to the existing AUROC =0.823, and the performance of IC is greater than EBGM in ADE detection.

Identifying the unintended effects of drugs (side effects) is a very important issue in pharmacological studies. The laboratory verification of associations between drugs and side effects requires costly, time-intensive research. Thus, an approach to predicting drug side effects based on known side effects, using a computational model, is highly desirable. To provide such a model, we used openly available data resources to model drugs and side effects as a bipartite graph. The drug-drug network is constructed using the word2vec model where the edges between drugs represent the semantic similarity between them. We integrated the bipartite graph and the semantic similarity graph using a matrix factorization method and a diffusion based model. Our results show the effectiveness of this integration by computing weighted (i.e., ranked) predictions of initially unknown links between side effects and drugs.

Identifying new relations between medical entities, such as drugs, diseases, and side effects, is typically a resource-intensive task, involving experimentation and clinical trials. The increased availability of related data and curated knowledge enables a computational approach to this task, notably by training models to predict likely relations. Such models rely on meaningful representations of the medical entities being studied. We propose a generic features vector representation that leverages co-occurrences of medical terms, linked with PubMed citations. We demonstrate the usefulness of the proposed representation by inferring two types of relations: a drug causes a side effect and a drug treats an indication. To predict these relations and assess their effectiveness, we applied 2 modeling approaches: multi-task modeling using neural networks and single-task modeling based on gradient boosting machines and logistic regression. These trained models, which predict either side effects or indications, obtained significantly better results than baseline models that use a single direct co-occurrence feature. The results demonstrate the advantage of a comprehensive representation. Selecting the appropriate representation has an immense impact on the predictive performance of machine learning models. Our proposed representation is powerful, as it spans multiple medical domains and can be used to predict a wide range of relation types. The discovery of new relations between various medical entities can be translated into meaningful insights, for example, related to drug development or disease understanding. Our representation of medical entities can be used to train models that predict such relations, thus accelerating healthcare-related discoveries.