Biomedical Big Data Research Articles

Evaluating deep neural networks in optimizing drug discovery and precision medicine: A review

Introduction/Background: Deep neural networks have shown great promise in advancing drug discovery and precision medicine. By leveraging large amounts of complex biomedical and chemical data, deep learning approaches can identify novel targets, predict drug-target and drug-drug interactions, generate new molecular structures, and assist in personalized treatment selection and development. However, fully utilizing deep learning techniques for optimization across the drug development pipeline remains an ongoing challenge. Materials and Methods: A comprehensive literature review was conducted using major bibliographic databases including PubMed, Web of Science, and Scopus. Search terms included combinations of "deep learning", "drug discovery", "precision medicine", "biomedical data", and "neural networks". Over 200 papers published between 2010-2023 related to deep learning applications in pharmacology and genomics were identified and reviewed. Results: Deep learning has been widely applied at various stages of the drug discovery process including target identification/prioritization, lead generation/optimization, and prediction of molecular properties. Convolutional neural networks are commonly used for the representation and classification of biological sequence and image data for tasks such as gene expression analysis and pathogen detection from microscopy images. Graph neural networks effectively model compound structures and interactome networks to predict molecular bindings and disease associations. Multi-modal neural networks integrate diverse data types for personalized treatment response prediction and biomarker discovery. Challenges remain around data and model interpretation, generalization to new targets/diseases, and integration across domains. Discussion: While deep learning has shown promise, rigorous benchmarking and validation on real-world clinical endpoints are still needed to establish usefulness in decision-making. Data and model transparency must be improved to enable scientific insights. Privacy and security risks accompanying "real world" biomedical big data will require ethical practices. Standardization and sharing of resources/protocols could accelerate progress by enabling comparison of techniques. Combining deep learning with other AI paradigms like causal inference may further improve utility in drug discovery and precision healthcare. Conclusion: Deep neural networks demonstrate potential for optimizing drug development and precision medicine applications. Continued advancement relies on addressing challenges around data, models, validation, and ethics. Multi-disciplinary collaborations integrating machine learning, molecular biology, medicine, and other domains are needed to fully realize benefits to patients.

Revealing Comprehensive Food Functionalities and Mechanisms of Action through Machine Learning.

Foods possess a range of unexplored functionalities; however, fully identifying these functions through empirical means presents significant challenges. In this study, we have proposed an in silico approach to comprehensively predict the functionalities of foods, encompassing even processed foods. This prediction is accomplished through the utilization of machine learning on biomedical big data. Our focus revolves around disease-related protein pathways, wherein we statistically evaluate how the constituent compounds collaboratively regulate these pathways. The proposed method has been employed across 876 foods and 83 diseases, leading to an extensive revelation of both food functionalities and their underlying operational mechanisms. Additionally, this approach identifies food combinations that potentially affect molecular pathways based on interrelationships between food functions within disease-related pathways. Our proposed method holds potential for advancing preventive healthcare.

Guest Editorial XAI Based Biomedical Big Data Privacy and Security

Guest Editorial XAI Based Biomedical Big Data Privacy and Security

Development and External Validation of Machine Learning Models for Diabetic Microvascular Complications: Cross-Sectional Study With Metabolites.

Diabetic kidney disease (DKD) and diabetic retinopathy (DR) are major diabetic microvascular complications, contributing significantly to morbidity, disability, and mortality worldwide. The kidney and the eye, having similar microvascular structures and physiological and pathogenic features, may experience similar metabolic changes in diabetes. This study aimed to use machine learning (ML) methods integrated with metabolic data to identify biomarkers associated with DKD and DR in a multiethnic Asian population with diabetes, as well as to improve the performance of DKD and DR detection models beyond traditional risk factors. We used ML algorithms (logistic regression [LR] with Least Absolute Shrinkage and Selection Operator and gradient-boosting decision tree) to analyze 2772 adults with diabetes from the Singapore Epidemiology of Eye Diseases study, a population-based cross-sectional study conducted in Singapore (2004-2011). From 220 circulating metabolites and 19 risk factors, we selected the most important variables associated with DKD (defined as an estimated glomerular filtration rate <60 mL/min/1.73 m2) and DR (defined as an Early Treatment Diabetic Retinopathy Study severity level ≥20). DKD and DR detection models were developed based on the variable selection results and externally validated on a sample of 5843 participants with diabetes from the UK biobank (2007-2010). Machine-learned model performance (area under the receiver operating characteristic curve [AUC] with 95% CI, sensitivity, and specificity) was compared to that of traditional LR adjusted for age, sex, diabetes duration, hemoglobin A1c, systolic blood pressure, and BMI. Singapore Epidemiology of Eye Diseases participants had a median age of 61.7 (IQR 53.5-69.4) years, with 49.1% (1361/2772) being women, 20.2% (555/2753) having DKD, and 25.4% (685/2693) having DR. UK biobank participants had a median age of 61.0 (IQR 55.0-65.0) years, with 35.8% (2090/5843) being women, 6.7% (374/5570) having DKD, and 6.1% (355/5843) having DR. The ML algorithms identified diabetes duration, insulin usage, age, and tyrosine as the most important factors of both DKD and DR. DKD was additionally associated with cardiovascular disease history, antihypertensive medication use, and 3 metabolites (lactate, citrate, and cholesterol esters to total lipids ratio in intermediate-density lipoprotein), while DR was additionally associated with hemoglobin A1c, blood glucose, pulse pressure, and alanine. Machine-learned models for DKD and DR detection outperformed traditional LR models in both internal (AUC 0.838 vs 0.743 for DKD and 0.790 vs 0.764 for DR) and external validation (AUC 0.791 vs 0.691 for DKD and 0.778 vs 0.760 for DR). This study highlighted diabetes duration, insulin usage, age, and circulating tyrosine as important factors in detecting DKD and DR. The integration of ML with biomedical big data enables biomarker discovery and improves disease detection beyond traditional risk factors.

Big biomedical data analytics in support of precision medicine

The goal of this paper is to present the major outcomes of the author scientific research in the area of Big Biomedical Data analytics in support of precision medicine, specifically genomics and oncology. The main parts of the paper reveal the state-of-the art in big data technologies and their significance for the advance of molecular biology and medicine in the context of the computational paradigm for scientific research. Furthermore, it presents approaches for gene finding and early detection of breast cancer and thyroid cancer based on big biomedical data analytics.

Call for Papers: Special Issue on Challenges and Opportunities in Biomedical Big Data Analysis: From Large Language Models to Clinical Applications

Call for Papers: Special Issue on Challenges and Opportunities in Biomedical Big Data Analysis: From Large Language Models to Clinical Applications

Artificial Intelligence with Great Potential in Medical Informatics: A Brief Review

In the 1950s and 1960s, in molecular biology, information technology was mainly applied to the molecular evolution of proteins and DNA, and later expanded to multiple fields such as sequence alignment, protein structure prediction, and gene splicing. Entering the 21st century, the completion of the Human Genome Project marks the arrival of the era of biomedical big data, providing a large amount of data for the application of artificial intelligence in this field. Especially in recent years, the continuous accumulation of medical data has pushed the application of artificial intelligence in the medical field to a broader and more practical level. This paper briefly introduces the applications of artificial intelligence in genomics, proteomics, transcriptomics, epigenetics, drug development, and other fields. I hope this review can clearly introduce which biomedical fields artificial intelligence can be applied to, and also promote doctors and related scholars to actively use artificial intelligence technology to solve specific biomedical problems.

A computational statistical approach to assess cowpea [Vigna unguiculata (L.) Walp.] cultivars diversity and select elite genotypes from Agro-morphological and biochemical traits

Computational statistic approaches play an essential role in the evaluation and processing of agronomic, biological, and bio-medical big data. The complexity and large size of those data make computational statistics a crucial tool in bio-statistical analysis. Based on this evidence, we characterized phenotype performances of cowpea cultivar in the Northern of Côte d’Ivoire, developing our own computational statistical approach, exclusively in the R programming language. Several packages of R have been executed to assess cowpea cultivar agro-morphological and biochemical performances. Z-score clustering analysis revealed four groups of cowpeas based on agro-morphological parameters. K-mean clustering survey revealed four and two groups of cowpea cultivar respectively for agro-morphological and biochemical parameters. The Horn parallel analysis highlighted two, four, two and two agro-explanatory components and/or agro-morphological parameters as influencing data variability respectively in the first, second, third, and fourth groups of cowpea cultivars previously revealed by the k-mean analysis. The same analysis exhibited two components in terms of biochemical parameters, inducing the variability in the two cluster groups of cowpea cultivar revealed by the k-mean survey. Integrative analysis of the ANOVA test and Tukey’s multi-comparative analysis displayed yield (agro-morphological parameter) and cowpea energetic content (biochemical parameter) as main sources of cowpea cultivar phenotypical variability (P &lt; 0.05). Of note, receiver operating characteristic predictive analysis showed the excellent performance of both cowpea yield and energetic content in selecting cowpea genetic germplasm area under the curve (AUC = 0.9). Considering as a whole, the present computational statistical approach shows excellent performance in the evaluation, characterization, and management of agro-morphological (yield parameter) and biochemical (energetic content) features of cowpea in genetic selection procedures.

Biomedical Big Data Technologies, Applications, and Challenges for Precision Medicine: A Review.

The explosive growth of biomedical Big Data presents both significant opportunities and challenges in the realm of knowledge discovery and translational applications within precision medicine. Efficient management, analysis, and interpretation of big data can pave the way for groundbreaking advancements in precision medicine. However, the unprecedented strides in the automated collection of large-scale molecular and clinical data have also introduced formidable challenges in terms of data analysis and interpretation, necessitating the development of novel computational approaches. Some potential challenges include the curse of dimensionality, data heterogeneity, missing data, class imbalance, and scalability issues. This overview article focuses on the recent progress and breakthroughs in the application of big data within precision medicine. Key aspects are summarized, including content, data sources, technologies, tools, challenges, and existing gaps. Nine fields-Datawarehouse and data management, electronic medical record, biomedical imaging informatics, Artificial intelligence-aided surgical design and surgery optimization, omics data, health monitoring data, knowledge graph, public health informatics, and security and privacy-are discussed.

Applications of Big Data and AI-Driven Technologies in CADD (Computer-Aided Drug Design).

In the field of computer-aided drug design (CADD), there has been dramatic progress in the development of big data and AI-driven methodologies. The expensive and time-consuming process of drug design is related to biomedical complexity. CADD can be used to apply effective and efficient strategies to overcome obstacles in the field of drug design in order to properly design and develop a new medicine. To prepare the raw data for consistent and repeatable applications of big data and AI methodologies, data pre-processing methods are introduced. Big data and AI technologies can be used to develop drugs in areas including predicting absorption, distribution, metabolism, excretion, and toxicity properties as well as finding binding sites in target proteins and conducting structure-based virtual screenings. The accurate and thorough analysis of large amounts of biomedical data as well as the design of prediction models in the area of drug design is made possible by data pre-processing and applications of big data and AI skills. In the biomedical big data era, knowledge on the biological, chemical, or pharmacological structures of biomedical entities relevant to drug design should be analyzed with significant big data and AI approaches.

Heart rate estimation from ballistocardiogram signals processing via low-cost telemedicine architectures: a comparative performance evaluation.

Medical devices (MDs) have been designed for monitoring the parameters of patients in many sectors. Nonetheless, despite being high-performing and reliable, they often turn out to be expensive and intrusive. In addition, MDs are almost exclusively used in controlled, hospital-based environments. Paving a path of technological innovation in the clinical field, a very active line of research is currently dealing with the possibility to rely on non-medical-graded low-cost devices, to develop unattended telemedicine (TM) solutions aimed at non-invasively gathering data, signals, and images. In this article, a TM solution is proposed for monitoring the heart rate (HR) of patients during sleep. A remote patient monitoring system (RPMS) featuring a smart belt equipped with pressure sensors for ballistocardiogram (BCG) signals sampling was deployed. A field trial was then conducted over a 2-month period on 24 volunteers, who also agreed to wear a finger pulse oximeter capable of producing a photoplethysmography (PPG) signal as the gold standard, to examine the feasibility of the solution via the estimation of HR values from the collected BCG signals. For this purpose, two of the highest-performing approaches for HR estimation from BCG signals, one algorithmic and the other based on a convolutional neural network (CNN), were retrieved from the literature and updated for a TM-related use case. Finally, HR estimation performances were assessed in terms of patient-wise mean absolute error (MAE). Results retrieved from the literature (controlled environment) outperformed those achieved in the experimentation (TM environment) by 29% (MAE = 4.24 vs. 5.46, algorithmic approach) and 52% (MAE = 2.32 vs. 3.54, CNN-based approach), respectively. Nonetheless, a low packet loss ratio, restrained elaboration time of the collected biomedical big data, low-cost deployment, and positive feedback from the users, demonstrate the robustness, reliability, and applicability of the proposed TM solution. In light of this, further steps will be planned to fulfill new targets, such as evaluation of respiratory rate (RR), and pattern assessment of the movement of the participants overnight.

Latest Findings in Key Research Areas of Precision Nursing for Chronic Diseases in Older Adults

The advent of the era of biomedical big data has helped promote the development of precision nursing. Precision nursing for chronic diseases in older adults is an interdisciplinary research field in which accurate individualized data are utilized to carry out early screening and health management of older adult populations at high risk for chronic diseases and early intervention of diseases, which plays an important role in improving the prognosis of diseases and the health level of the older adult population. Herein, we introduced the concept of precision nursing, and discussed the latest research findings in the key areas of precision nursing for chronic diseases in older adults, including precision symptom management in cancer patients and precision nursing in older patients with multimorbidity. At present, research concerning precise symptom management of cancer patients is mainly focused on prediction modelling for risks of symptoms, longitudinal change trajectories, core symptom identification, etc. Investigations in the precise nursing of cancer patients are conducted in the following areas, risk prediction, the timing of interventions, and intervention targets. Research on precision nursing for multimorbidity is mainly focused on assessment of chronic disease multimorbidity, multimorbidity pattern recognition, and health management of multimorbidity. We also discussed potential opportunities and challenges of precision nursing in the future, in order to provide a scientific basis for the improving the practice and theories of precision nursing. In the future, precision nursing will play an ever more important role in uncovering pathogenic information, the diagnosis and treatment of diseases, the health of the research population, and the promotion of medical research.

Investigating Employees’ Concerns and Wishes Regarding Digital Stress Management Interventions With Value Sensitive Design: Mixed Methods Study

Work stress places a heavy economic and disease burden on society. Recent technological advances include digital health interventions for helping employees prevent and manage their stress at work effectively. Although such digital solutions come with an array of ethical risks, especially if they involve biomedical big data, the incorporation of employees' values in their design and deployment has been widely overlooked. To bridge this gap, we used the value sensitive design (VSD) framework to identify relevant values concerning a digital stress management intervention (dSMI) at the workplace, assess how users comprehend these values, and derive specific requirements for an ethics-informed design of dSMIs. VSD is a theoretically grounded framework that front-loads ethics by accounting for values throughout the design process of a technology. We conducted a literature search to identify relevant values of dSMIs at the workplace. To understand how potential users comprehend these values and derive design requirements, we conducted a web-based study that contained closed and open questions with employees of a Swiss company, allowing both quantitative and qualitative analyses. The values health and well-being, privacy, autonomy, accountability, and identity were identified through our literature search. Statistical analysis of 170 responses from the web-based study revealed that the intention to use and perceived usefulness of a dSMI were moderate to high. Employees' moderate to high health and well-being concerns included worries that a dSMI would not be effective or would even amplify their stress levels. Privacy concerns were also rated on the higher end of the score range, whereas concerns regarding autonomy, accountability, and identity were rated lower. Moreover, a personalized dSMI with a monitoring system involving a machine learning-based analysis of data led to significantly higher privacy (P=.009) and accountability concerns (P=.04) than a dSMI without a monitoring system. In addition, integrability, user-friendliness, and digital independence emerged as novel values from the qualitative analysis of 85 text responses. Although most surveyed employees were willing to use a dSMI at the workplace, there were considerable health and well-being concerns with regard to effectiveness and problem perpetuation. For a minority of employees who value digital independence, a nondigital offer might be more suitable. In terms of the type of dSMI, privacy and accountability concerns must be particularly well addressed if a machine learning-based monitoring component is included. To help mitigate these concerns, we propose specific requirements to support the VSD of a dSMI at the workplace. The results of this work and our research protocol will inform future research on VSD-based interventions and further advance the integration of ethics in digital health.

3PNMF-MKL: A non-negative matrix factorization-based multiple kernel learning method for multi-modal data integration and its application to gene signature detection.

In this current era, biomedical big data handling is a challenging task. Interestingly, the integration of multi-modal data, followed by significant feature mining (gene signature detection), becomes a daunting task. Remembering this, here, we proposed a novel framework, namely, three-factor penalized, non-negative matrix factorization-based multiple kernel learning with soft margin hinge loss (3PNMF-MKL) for multi-modal data integration, followed by gene signature detection. In brief, limma, employing the empirical Bayes statistics, was initially applied to each individual molecular profile, and the statistically significant features were extracted, which was followed by the three-factor penalized non-negative matrix factorization method used for data/matrix fusion using the reduced feature sets. Multiple kernel learning models with soft margin hinge loss had been deployed to estimate average accuracy scores and the area under the curve (AUC). Gene modules had been identified by the consecutive analysis of average linkage clustering and dynamic tree cut. The best module containing the highest correlation was considered the potential gene signature. We utilized an acute myeloid leukemia cancer dataset from The Cancer Genome Atlas (TCGA) repository containing five molecular profiles. Our algorithm generated a 50-gene signature that achieved a high classification AUC score (viz., 0.827). We explored the functions of signature genes using pathway and Gene Ontology (GO) databases. Our method outperformed the state-of-the-art methods in terms of computing AUC. Furthermore, we included some comparative studies with other related methods to enhance the acceptability of our method. Finally, it can be notified that our algorithm can be applied to any multi-modal dataset for data integration, followed by gene module discovery.

Prediction via Similarity: Biomedical Big Data and the Case of Cancer Models

In recent years, the biomedical field has witnessed the emergence of novel tools and modelling techniques driven by the rise of the so-called Big Data. In this paper, we address the issue of predictability in biomedical Big Data models of cancer patients, with the aim of determining the extent to which computationally driven predictions can be implemented by medical doctors in their clinical practice. We show that for a specific class of approaches, called k-Nearest Neighbour algorithms, the ability to draw predictive inferences relies on a geometrical, or topological, notion of similarity encoded in a well-defined metric, which determines how close the characteristics of distinct patients are on average. We then discuss the conditions under which the relevant models can yield reliable and trustworthy predictive outcomes.

Predictive health intelligence: Potential, limitations and sense making.

We discuss the new paradigm of predictive health intelligence, based on the use of modern deep learning algorithms and big biomedical data, along the various dimensions of: a) its potential, b) the limitations it encounters, and c) the sense it makes. We conclude by reasoning on the idea that viewing data as the unique source of sanitary knowledge, fully abstracting from human medical reasoning, may affect the scientific credibility of health predictions.

Prediction of protein-protein interactions based on ensemble residual convolutional neural network

With the development of bio-medical big data, the prediction of protein-protein interactions (PPIs) with the help of deep learning (DL) has attracted much attention for the study of intermolecular mechanism, drug design, human disease treatment. Given that the experiment-based methods can be difficult, reliable and DL-based approaches are needed. In this paper, we develop the EResCNN, an effective predictor to predict PPIs based on ensemble residual convolutional neural network. First, the fused feature representation is captured by concatenating the vectors obtained via pseudo amino acid composition (PseAAC), auto covariance descriptor (AC), pseudo position-specific scoring matrix (PsePSSM), encoding based on grouped weight (EBGW), multivariate mutual information (MMI) and conjoint triad (CT). Then the high-level information can be obtained using the convolution and pooling of the residual convolutional neural network (RCNN) via layer-by-layer learning. At last, we ensemble RCNN, XGBoost, random forest, LightGBM and extremely randomized trees to build the EResCNN model. The predictive results indicate that EResCNN achieves better performance with the ACC values of 95.34%, 87.89% and 98.61% on S. cerevisiae, H. pylori and Human-Y. pestis datasets, respectively. We also apply EResCNN to the datasets of H. sapiens, M. musculus, C. elegans and E. coli for cross-species prediction. Especially, we find EResCNN can infer significant PPIs network on one-core network, Wnt-related signal pathway network, cancer-specific network and multi-core network, which could provide some references for signal pathway research, disease-related gene mining, and interaction network topology.

Exploratory factor analysis determines latent factors in Guillain–Barré syndrome

Exploratory factor analysis (EFA) has been developed as a powerful statistical procedure in psychological research. EFA’s purpose is to identify the nature and number of latent constructs (= factors) underlying a set of observed variables. Since the research goal of EFA is to determine what causes the observed responses, EFA is ideal for hypothesis-based studies, such as identifying the number and nature of latent factors (e.g., cause, risk factors, etc.). However, the application of EFA in the biomedical field has been limited. Guillain–Barré syndrome (GBS) is peripheral neuropathy, in which the presence of antibodies to glycolipids has been associated with clinical signs. Although the precise mechanism for the generation of anti-glycolipid antibodies is unclear, we hypothesized that latent factors, such as distinct autoantigens and microbes, could induce different sets of anti-glycolipid antibodies in subsets of GBS patients. Using 55 glycolipid antibody titers from 100 GBS and 30 control sera obtained by glycoarray, we conducted EFA and extracted four factors related to neuroantigens and one potentially suppressive factor, each of which was composed of the distinct set of anti-glycolipid antibodies. The four groups of anti-glycolipid antibodies categorized by unsupervised EFA were consistent with experimental and clinical findings reported previously. Therefore, we proved that unsupervised EFA could be applied to biomedical data to extract latent factors. Applying EFA for other biomedical big data may elucidate latent factors of other diseases with unknown causes or suppressing/exacerbating factors, including COVID-19.

Hyper-parameter optimization of multiple machine learning algorithms for molecular property prediction using hyperopt library

Due to outstanding performance in cheminformatics, machine learning algorithms have been increasingly used to mine molecular properties and biomedical big data. The performance of machine learning models is known to critically depend on the selection of the hyper-parameter configuration. However, many studies either explored the optimal hyper-parameters per the grid searching method or employed arbitrarily selected hyper-parameters, which can easily lead to achieving a suboptimal hyper-parameter configuration. In this study, Hyperopt Library embedding with the Bayesian optimization is employed to find optimal hyper-parameters for different machine learning algorithms. Six drug discovery datasets, including solubility, probe-likeness, hERG, Chagas disease, tuberculosis, and malaria, are used to compare different machine learning algorithms with ECFP6 fingerprints. This contribution aims to evaluate whether the Bernoulli Naïve Bayes, Logistic Linear Regression, AdaBoost Decision Tree, Random Forest, support vector machine, and deep neural networks algorithms with optimized hyper-parameters can offer any improvement in testing as compared with the Referenced Models assessed by an array of metrics including AUC, F1-score, Cohen’s kappa, Matthews correlation coefficient, recall, precision, and accuracy. Based on the rank normalized score approach, the Hyperopt Models achieve better or comparable performance on 33 out 36 models for different drug discovery datasets, showing significant improvement achieved by employing the Hyperopt library. The open-source code of all the 6 machine learning frameworks employed in the Hyperopt Python package is provided to make this approach accessible to more scientists, who are not familiar with writing code.

Integrative Learning of Structured High-Dimensional Data from Multiple Datasets.

Integrative learning of multiple datasets has the potential to mitigate the challenge of small n and large p that is often encountered in analysis of big biomedical data such as genomics data. Detection of weak yet important signals can be enhanced by jointly selecting features for all datasets. However, the set of important features may not always be the same across all datasets. Although some existing integrative learning methods allow heterogeneous sparsity structure where a subset of datasets can have zero coefficients for some selected features, they tend to yield reduced efficiency, reinstating the problem of losing weak important signals. We propose a new integrative learning approach which can not only aggregate important signals well in homogeneous sparsity structure, but also substantially alleviate the problem of losing weak important signals in heterogeneous sparsity structure. Our approach exploits a priori known graphical structure of features and encourages joint selection of features that are connected in the graph. Integrating such prior information over multiple datasets enhances the power, while also accounting for the heterogeneity across datasets. Theoretical properties of the proposed method are investigated. We also demonstrate the limitations of existing approaches and the superiority of our method using a simulation study and analysis of gene expression data from ADNI.