Mapping Drug Responses Through Multi-Omics: A New Era of Bioinformatics in Precision Medicine

Gargi Bachhav, Samruddhi Avhad, Janhavi Bhalerao, Mukund Pache, Akshat Bhamare, Sakshi Ahirrao,

doi:10.5281/zenodo.16914606

Review Paper | Open Access
Volume 02 | Issue 08 | Article Id IJSRT/250308033

Mapping Drug Responses Through Multi-Omics: A New Era of Bioinformatics in Precision Medicine
Gargi Bachhav* Samruddhi Avhad Janhavi Bhalerao Mukund Pache Akshat Bhamare Sakshi Ahirrao
Department of Pharmacology, K. V. N. Naik S. P. Sanstha’s, Institute of Pharmaceutical Education & Research, Canada Corner, Nashik, 422002, Maharashtra, India

Abstract

The advent of precision medicine has instigated a paradigm shift in disease treatment, emphasizing the customization of therapies based on individual molecular profiles. Central to this transformation is the integration of multi-omics data encompassing genomics, transcriptomics, proteomics, metabolomics, epigenomics, and more to elucidate the complex biological networks that influence drug response. This review synthesizes the current landscape of multi-omics approaches in precision pharmacology, concentrating on their application in understanding and predicting variability in therapeutic efficacy and toxicity. We commence by examining the individual omics layers pertinent to pharmacodynamics and pharmacokinetics, highlighting how their integration yields synergistic insights that surpass the limitations of single-omics strategies. The challenges posed by heterogeneous data types, technical variability, and integration complexity are critically assessed. In response, we evaluate computational methodologies developed to manage and analyse multi-omics datasets, including advanced machine learning and AI-driven platforms. Tools such as iCluster, MOFA, and DeepMO are discussed for their role in enhancing the accuracy of drug response predictions. We underscore key case studies demonstrating the clinical impact of multi-omics, such as in EGFR-targeted therapy and BRCA-mutated cancers, and explore emerging areas like the role of the microbiome and single-cell multi-omics in revealing intra-tumour heterogeneity. Finally, we address ethical considerations, data privacy, and the necessity for standardization. This review envisions a future where multi-omics-guided precision medicine optimizes drug efficacy and democratizes personalized care across diverse populations.

Keywords

Multi-omics, drug response, precision medicine, data integration, pharmacogenomics, machine learning, single-cell analysis

Introduction

Precision medicine, a transformative shift in healthcare, aspires to tailor medical treatment to each patient's individual characteristics. Central to this approach is the understanding that drug responses are not universally predictable but are modulated by a constellation of genetic, molecular, environmental, and lifestyle factors. Despite the promise of “the right drug for the right patient,” inter-individual variability in drug efficacy and adverse reactions continues to challenge clinicians. Bridging this knowledge gap necessitates a deeper, more holistic exploration of the biological determinants of therapeutic response[1]. Enter the era of multi-omics, an integrative framework that harnesses diverse layers of molecular information genomics, transcriptomics, proteomics, metabolomics, and epigenomics to provide a comprehensive view of biological systems. Each omics layer captures a distinct facet of cellular function: DNA mutations and structural variants (genomics), mRNA expression profiles (transcriptomics), protein abundance and modifications (proteomics), metabolite fluxes (metabolomics), and regulatory changes such as DNA methylation or histone modification (epigenomics). Together, they constitute a high-resolution map of the molecular landscape influencing drug behaviour[2]. While each individual omics modality has yielded valuable insights, the integration of these data types holds the greatest promise for understanding drug mechanisms, identifying biomarkers of response or resistance, and uncovering new therapeutic targets. This multidimensional approach enables the dissection of complex drug response phenotypes that single-layer analyses often oversimplify or overlook[3]. In recent years, technological advancements have exponentially increased the throughput and affordability of omics profiling, enabling the generation of massive datasets across populations and disease states. However, these developments have also introduced computational and interpretive challenges. Multi-omics data are inherently high-dimensional, noisy, and heterogeneous, requiring sophisticated algorithms and robust statistical frameworks for effective integration and interpretation[4]. This review aims to critically evaluate the current state of multi-omics in drug response mapping within the context of precision medicine. We begin by dissecting the unique contributions of each omics layer to pharmacological understanding and then delve into the challenges of integrating these diverse data sources. We examine computational strategies—ranging from early data concatenation models to advanced AI and network-based frameworks—that facilitate the extraction of meaningful patterns from complex datasets. In addition, we explore the emerging applications of multi-omics in precision oncology, pharmacogenomics, microbiome research, and single-cell biology, showcasing case studies where integrative omics has led to actionable clinical insights. We also address the ethical, legal, and social considerations intrinsic to handling sensitive multi-omics data, particularly in the context of data sharing and patient privacy. Ultimately, this review aspires to provide a comprehensive, yet critical, examination of how multi-omics reshapes drug development and therapy personalisation. By highlighting current innovations and unmet needs, we aim to chart a roadmap toward a future where multi-omics enhances our mechanistic understanding of drugs and accelerates the delivery of safer, more effective treatments for all patients.

Overview of Omics Layers Relevant to Drug Response

Understanding the molecular determinants of drug response requires dissecting multiple layers of biological regulation. Each omics layer contributes unique insights into how drugs interact with the human system. When integrated, these layers offer a synergistic and mechanistic understanding that can drive personalised treatment strategies.

Genomics: Blueprint of Drug Response

Genomic variation plays a foundational role in drug efficacy and toxicity. Germline polymorphisms in drug-metabolising enzymes, such as CYP2C9, CYP2D6, and TPMT, can significantly affect pharmacokinetics, leading to overexposure or underexposure to therapeutic agents. For example, polymorphisms in CYP2C19 influence the activation of clopidogrel, an antiplatelet agent, resulting in poor therapeutic outcomes in specific genotypes[5]. In oncology, somatic mutations in genes like EGFR, KRAS, or BRAF have emerged as predictive biomarkers for targeted therapies. For instance, non-small cell lung cancer patients harbouring EGFR mutations show robust responses to tyrosine kinase inhibitors such as erlotinib and gefitinib. Thus, genomic profiling has become a clinical cornerstone in cancer precision medicine[6].

Transcriptomics: Functional Readout of Gene Activity

Transcriptomics provides a dynamic snapshot of gene expression under specific physiological or pathological conditions. Unlike the static nature of the genome, transcriptomic changes can reflect immediate cellular responses to drugs. Pharmacotranscriptomics studies have revealed how expression changes in transporter or efflux pump genes, like ABCB1 or SLCO1B1, can modulate drug distribution and clearance[7]. Gene expression profiles have also been used to classify responders versus non-responders to chemotherapy. The development of the Oncotype DX and MammaPrint assays, which predict breast cancer recurrence risk and chemotherapy benefit, underscores the translational power of transcriptomic data[8].

Proteomics: Bridging Genotype and Phenotype

Proteins, the primary executors of cellular function, represent direct drug targets and mediators of pharmacologic effects. Quantitative proteomics measures protein abundance, turnover, and post-translational modifications (PTMs) such as phosphorylation or ubiquitination—key drug activity and resistance regulators[9]. For example, phosphorylation of the estrogen receptor can influence sensitivity to hormonal therapies in breast cancer. Mass spectrometry-based proteomics has also identified differential protein expression patterns that correlate with cisplatin resistance in ovarian cancer, facilitating predictive biomarker discovery[10].

Metabolomics: Snapshot of Biochemical Activity

Metabolomics captures the small-molecule metabolites that serve as substrates, intermediates, and products of cellular metabolism. Since drugs are metabolised into active or inactive compounds, metabolomics directly informs on drug bioavailability and toxicity[11]. For instance, metabolomic profiling has identified specific metabolic signatures associated with doxorubicin-induced cardiotoxicity, potentially allowing for early risk stratification. Additionally, changes in tryptophan metabolism have been implicated in immune checkpoint therapy responses, highlighting metabolomics as a tool for immunotherapy optimisation[12].

Epigenomics: Regulatory Modulators of Drug Response

Epigenetic modifications such as DNA methylation and histone acetylation dynamically regulate gene expression without altering the DNA sequence. Aberrant methylation of promoter regions can silence tumour suppressor genes, contributing to drug resistance. For example, hypermethylation of the MGMT gene is associated with increased sensitivity to temozolomide in glioblastoma[13]. Epigenomic drugs like DNA methyltransferase inhibitors (e.g., azacitidine) are now used to reverse resistant phenotypes in hematologic malignancies, demonstrating the therapeutic potential of targeting the epigenome[14].

Synergy Across Omics Layers:

Individually, each omics domain offers partial insights into drug response. However, these data layers can reveal complex regulatory networks that control cellular fate under pharmacologic intervention. For instance, integrating epigenomic and transcriptomic data can clarify whether gene silencing arises from DNA methylation or transcriptional repression. Similarly, linking proteomic and metabolomic data can illuminate how altered enzyme levels affect drug metabolism[15]. Thus, multi-omics integration represents a paradigm shift from single-biomarker models to systems-level therapeutic efficacy and safety predictors.

Figure 1. An integrated multi-omics framework for precision medicine. Bulk tumour samples are profiled across multiple biological layers—genomics, transcriptomics, proteomics, epigenomics, immunomics, and clinical data.

Table 1 Key Omics Layers and Their Role in Drug Response

Omics Layer	Key Insight	Example Application
Genomics	Genetic variants	CYP2C19 and clopidogrel metabolism
Transcriptomics	Gene expression changes	Chemotherapy response prediction
Proteomics	Protein abundance and PTMs	Cisplatin resistance profiling
Metabolomics	Drug metabolism products	Doxorubicin cardiotoxicity
Epigenomics	Gene regulation	MGMT methylation in glioblastoma

Challenges in Multi-Omics Data Integration

While the promise of multi-omics integration in drug response research is immense, translating this potential into clinical practice remains fraught with significant challenges. These barriers span technical, analytical, and biological domains, and addressing them is critical to unlocking the full utility of multi-layered data.

Data Heterogeneity and Format Incompatibility:

Omics technologies generate highly heterogeneous data regarding dimensionality, structure, and file formats. Genomic data are often categorical (e.g., variant presence), transcriptomic and proteomic data are continuous, while metabolomic and epigenomic datasets can be semi-quantitative. Integrating such diverse data types requires specialised frameworks to preserve biological meaning across modalities[16].

Scaling and Normalisation Issues:

Different omics layers operate on vastly different scales, ranging from transcript counts in thousands to metabolite concentrations in micromolar levels. Proper normalisation is crucial to ensure no single data type dominates the integrative analysis. However, harmonising these scales without introducing bias remains a persistent challenge[17].

Missing Data, Batch Effects, and Noise:

Missing data are common due to sample attrition, detection limits, or failed assays. If not corrected, batch effects—systematic differences arising from experimental or processing conditions can introduce spurious correlations. Noise from low-abundance features or instrument variability can further obscure accurate biological signals. Tools like ComBat or RUV have been developed to correct batch effects, but their efficacy varies across datasets and omics types[18].

Lack of Standardised Pipelines and Gold Standards:

Unlike genomic analysis, where variant-calling pipelines are well-established, multi-omics lacks universally accepted workflows. This leads to methodological inconsistency and poor reproducibility. Additionally, there are few benchmark datasets or gold standards for validating integration methods, limiting the ability to compare performance across algorithms objectively[19].

Biological Versus Technical Variability:

Disentangling biological variability (e.g., tumour heterogeneity) from technical artefacts is a non-trivial problem. For example, variability in gene expression may reflect proper differential regulation or stem from sample degradation. Integrative analyses must carefully control for confounders to avoid drawing incorrect conclusions[20].

Interpretability of Integration Models:

Interpretability becomes a concern as integrative models grow more complex, especially with the advent of deep learning. Clinically actionable insights require transparent models that trace outputs to specific molecular features. Current efforts in explainable AI (XAI) for multi-omics are still in their infancy[21].

Data Volume and Computational Burden:

Multi-omics analyses are data-intensive, often requiring high-performance computing and scalable storage solutions. Data processing and integration become resource-heavy tasks for large-scale studies involving thousands of patients, limiting accessibility for smaller research groups[22]. In summary, while multi-omics integration holds transformative potential, significant methodological and infrastructural hurdles must be overcome. Addressing these challenges will be essential for developing robust, reproducible, and clinically translatable models of drug response.

Computational Approaches for Data Integration and Analysis

Effective multi-omics data integration is essential for capturing the multifaceted biological processes underpinning drug response. The diversity of omics type’s structure, scale, and dimensionality necessitate sophisticated computational strategies. These can broadly be classified into early, intermediate, and late integration approaches, each offering distinct advantages and limitations.

Figure 2. Horizontal, vertical, and post-analysis strategies for multi-omics data integration to support precision medicine applications.

Table 2 Notable Multi-Omics Tools and Their Functions

Tool	Method Type	Application
iCluster	Intermediate integration	Subtype classification
MOFA	Joint factor analysis	Feature selection, clustering
DeepMO	Deep learning	Drug synergy prediction
SNF	Network fusion	Patient stratification

Early Integration: Concatenation-Based Models:

Early integration, or horizontal integration, involves merging raw or normalised data from different omics layers into a unified matrix. This concatenated matrix is then used for downstream analysis such as clustering, classification, or regression[23]. While simple to implement, early integration often suffers from overfitting due to the “curse of dimensionality,” where the number of features far exceeds the number of samples. Additionally, differences in data scale or noise distribution can obscure biological signals. Nevertheless, this approach remains foundational in multi-omics machine learning pipelines, particularly when dimensionality reduction techniques are applied[23].

Intermediate Integration: Joint Latent Representations:

Intermediate integration seeks to extract shared patterns from different omics layers by learning joint latent representations. One of this category's most widely used tools is iCluster, which employs a joint latent variable model to cluster multi-omics datasets, identifying coherent molecular subtypes simultaneously[24]. Another robust framework is MOFA (Multi-Omics Factor Analysis), which captures hidden factors driving variance across data types, allowing for interpretable integration and feature selection. mixOmics and its extension DIABLO offer supervised and unsupervised methods for integrative variable selection and classification, frequently used in biomarker discovery for drug response[25].

Late Integration: Decision-Level Fusion:

Late integration processes each omics layer independently, extracting model outputs or combined predictions in a second step. This strategy is beneficial when omics datasets are collected from non-overlapping cohorts or have incompatible scales. Ensemble learning or meta-analysis methods are often applied in this context, ensuring robustness and interpretability[23].

Network-Based Integration and Pathway Mapping:

Network-based approaches integrate multi-omics data by mapping them onto biological networks, such as protein-protein interaction (PPI) or gene regulatory networks. The Similarity Network Fusion (SNF) algorithm is a popular method that constructs similarity graphs for each omics type and then fuses them into a consensus network. SNF has been successfully applied to identify patient subgroups with distinct drug sensitivity profiles. Pathway-centric methods allow integration to be anchored in biological meaning. Tools like PARADIGM and Pathway-Express project omics data onto known signalling pathways, enabling functional interpretations that are more actionable in clinical contexts[26].

Dimensionality Reduction Techniques:

Given the high dimensionality of multi-omics data, dimensionality reduction is critical for visualisation, clustering, and machine learning. Principal component analysis (PCA) and t-distributed stochastic neighbour embedding (t-SNE) are commonly used but are often limited to single-omics applications.

Newer methods tailored for multi-omics include Canonical Correlation Analysis (CCA) and Multiple Co-Inertia Analysis (MCIA), which identify correlated features across data types. Non-linear approaches such as UMAP and autoencoders enhance feature compression while preserving data structure[27]. In sum, computational integration of multi-omics is evolving rapidly, with emerging methods enabling deeper, systems-level insights into drug response. Selection of the optimal integration strategy depends on the study goal, data characteristics, and biological interpretability.

AI & Machine Learning for Drug Response Prediction

Artificial intelligence (AI) and machine learning (ML) are indispensable tools in analysing multi-omics data to predict drug response. These methods can discern complex, non-linear relationships across diverse datasets, offering unprecedented precision in stratifying patients and predicting therapeutic efficacy[28].

Supervised Machine Learning Models:

Supervised ML models require labelled training data, where drug responses are known. Among the most commonly used algorithms are:

Support Vector Machines (SVMS): Effective in high-dimensional spaces and commonly used for binary classifying responders versus non-responders.

Random Forests: Ensemble models that aggregate predictions from multiple decision trees, providing robustness and feature importance ranking.

XGBoost: An optimised gradient boosting algorithm known for speed and performance in structured data, including pharmacogenomic studies. These models often use feature selection techniques to manage high-dimensional omics data, identifying the most predictive genes, proteins, or metabolites.

Deep Learning: Harnessing Hierarchical Features:

Deep learning models have revolutionised omics analysis by automatically learning hierarchical representations from complex input data.

Autoencoders compress multi-omics data into latent vectors that retain essential features. These are particularly useful for unsupervised learning and clustering.

Although initially developed for image recognition, Convolutional Neural Networks (CNNs) have been adapted to identify spatial and sequential patterns in omics data.

Graph Neural Networks (GNNs) model biological systems as graphs, capturing topological relationships among genes or proteins. They are beneficial for integrating omics data with PPI networks or signalling pathways. One example is Drug Cell, a biologically interpretable deep learning model that links genetic alterations to drug sensitivity by modelling gene-drug interactions through hierarchical cellular subsystems. DeepMO is another framework that integrates multi-omics via modality-specific neural networks, enabling improved prediction of drug synergy[29–32].

Multi-View Learning and Omics Integration:

Multi-view learning is an ML paradigm designed to handle multiple, complementary data views—ideal for multi-omics. By constructing separate learners for each omics type and then integrating predictions, multi-view models can preserve modality-specific signals while enhancing generalisation. Ensemble methods and attention mechanisms are often used in this context to dynamically weigh the contribution of each omics layer. OmicsNet, for example, enables multi-view learning with biological network integration, enhancing interpretability in drug target identification[33].

Feature Selection in High-Dimensional Spaces:

Due to the disproportionate ratio of features to samples in omics data, feature selection is essential to avoid overfitting and improve model interpretability. Techniques like LASSO, Recursive Feature Elimination (RFE), and tree-based feature importance scoring are widely used. Recent methods integrate biological prior knowledge, such as gene pathways or transcription factor networks, to guide feature selection, improving accuracy and clinical relevance[34].

Model Validation and Interpretability:

A key challenge remains the generalizability and interpretability of ML models in clinical settings. To ensure model robustness, cross-validation, bootstrapping, and external validation using independent cohorts are used. Moreover, explainable AI (XAI) tools such as SHAP (SHapley Additive Explanations) and LIME (Local Interpretable Model-Agnostic Explanations) are increasingly employed to elucidate how features drive predictions. In conclusion, AI and ML models are redefining the landscape of drug response prediction. By integrating multi-omics data into interpretable and accurate predictive frameworks, these technologies are paving the way toward more personalised and effective therapeutics[35] .

Case Studies in Precision Medicine & Drug Development

Real-world applications of multi-omics in drug response research have yielded critical breakthroughs in precision oncology. Large-scale initiatives such as The Cancer Genome Atlas (TCGA) and the International Cancer Genome Consortium (ICGC) have provided a wealth of multi-modal data, offering fertile ground for biomarker discovery and treatment stratification.

EGFR Inhibitors in Non-Small Cell Lung Cancer:

A landmark case study in precision medicine involves using tyrosine kinase inhibitors (TKIs) targeting EGFR mutations in non-small cell lung cancer (NSCLC). Early genomic analyses identified specific activating mutations such as exon 19 deletions and L858R substitutions associated with sensitivity to TKIs like gefitinib and erlotinib. Proteomic and transcriptomic data integration further refined patient stratification, identifying resistance mechanisms involving MET amplification and T790M secondary mutations. These insights now inform routine clinical testing and therapeutic decision-making[36].

BRCA Mutations and PARP Inhibitors:

Another success story involves BRCA1/2 mutations and poly (ADP-ribose) polymerase (PARP) inhibitors in breast and ovarian cancer. TCGA datasets revealed how homologous recombination deficiency, often caused by BRCA mutations, sensitises tumours to DNA-damaging agents. Multi-omics analyses integrating genomic instability scores, methylation data, and gene expression signatures have enabled the development of HRD scores, which guide PARP inhibitor use beyond BRCA-mutant tumours (The Cancer Genome Atlas Network, 2011)[37].

Pharmacogenomic Consortia and Clinical Trials:

The Genomics of Drug Sensitivity in Cancer (GDSC) and the Cancer Cell Line Encyclopaedia (CCLE) have systematically profiled hundreds of cell lines across thousands of drugs, linking genomic alterations to drug sensitivity phenotypes. These pharmacogenomic consortia have underpinned numerous multi-omics modelling efforts, validating predictive biomarkers such as NRAS mutations in MEK inhibitor response. Real-world clinical trials are increasingly integrating omics data for adaptive therapy selection. The NCI-MATCH trial uses genomic screening to assign patients to targeted therapies. In contrast, the WINTHER trial incorporates transcriptomic data to expand treatment matching, illustrating the clinical feasibility of multi-omics-guided therapy. These case studies exemplify how integrative omics has transitioned from bench to bedside, improving therapeutic precision and patient outcomes[38].

Microbiome in Drug Response and Integration with Host Omics

The gut microbiome plays a critical and increasingly recognised role in modulating drug response, primarily through its influence on drug metabolism, immune modulation, and systemic signalling pathways. Integrating microbiome data with host multi-omics is now emerging as a key area in precision pharmacology.

Microbiota-Drug Interactions:

Microbial enzymes can chemically modify drugs, affecting their bioavailability, efficacy, and toxicity. For instance, β-glucuronidase-producing bacteria in the gut re-activate the chemotherapeutic irinotecan, leading to gastrointestinal toxicity. Similarly, microbial metabolism of digoxin by Eggerthella lenta has been shown to inactivate the drug, explaining inter-patient variability in response. These findings underscore the importance of incorporating microbial enzymatic capacity into pharmacokinetic modelling and therapeutic planning[39].

Inter-Omic Approaches:

Researchers are increasingly turning to inter-omic integration to unravel the host-microbiome-drug interaction triad. Host transcriptomic and metabolomic profiling can reveal immune and metabolic pathways modulated by microbial metabolites. For example, microbial short-chain fatty acids (SCFAs) influence histone acetylation and immune checkpoint expression, affecting response to cancer immunotherapy.[40] Tools like PICRUSt and QIIME allow for microbial community profiling and metagenomic prediction, which can be aligned with host omics using frameworks such as multi-block partial least squares (MB-PLS) and network-based correlation analysis[41].

Clinical Implications:

In cancer immunotherapy, responders often exhibit distinct gut microbiome signatures enriched with Akkermansia muciniphila and Faecalibacterium prausnitzii. Faecal microbiota transplants (FMT) from responders into germ-free mice have been shown to restore checkpoint blockade sensitivity, providing functional validation[42]. These examples illustrate how integrating microbiome and host omics may improve predictive modelling and therapeutic stratification in clinical settings.

Figure 3. Host–microbiome–environment interactions influencing gene regulation and phenotype, including drug response.

Single-Cell Multi-Omics & Heterogeneity in Drug Response

Tumours and other complex tissues exhibit significant cellular heterogeneity, impacting drug efficacy and resistance. Traditional bulk omics approaches average signals across diverse cell populations, potentially masking critical subpopulation dynamics. Single-cell multi-omics technologies have emerged as transformative tools for resolving this heterogeneity at unprecedented resolution.

Single-Cell Multi-Omics Technologies:

Single-cell RNA sequencing (scRNA-seq) has revolutionised our understanding of cellular diversity by capturing transcriptomic profiles at single-cell resolution. Advances in multi-modal technologies, such as scRNA-seq + ATAC-seq (simultaneous chromatin accessibility and gene expression profiling), enable direct linkage between transcriptional states and regulatory landscapes. Similarly, CITE-seq combines surface protein quantification with transcriptomic data, bridging proteomics and gene expression at the single-cell level. These methods allow for identifying distinct cell states, rare subpopulations, and lineage hierarchies contributing to variable drug responses[43].

Spatial Omics Integration:

The spatial context of cells within tissues further influences their behaviour and drug sensitivity. Techniques such as spatial transcriptomics and imaging mass cytometry enable mapping of molecular features back to their histological locations. This spatially resolved omics data enhances our understanding of how microenvironmental niches contribute to therapy resistance, particularly in the tumour microenvironment (TME). For example, spatial transcriptomic profiling of melanoma tumours has revealed immunosuppressive niches resistant to checkpoint blockade, guiding targeted combination therapy strategies[44, 45].

Figure 4. Spatially resolved single-cell multi-omics analysis of tumour microenvironments reveals resistance and response patterns in immunotherapy-treated patients.

Pharmacogenomics and Personalised Drug Therapy

Pharmacogenomics—the study of how genetic variation influences drug response—is at personalised medicine's core. By identifying genetic polymorphisms that affect drug metabolism, efficacy, or toxicity, pharmacogenomics enables the development of safer and more effective individualised therapies.

Genetic Variants and Drug Metabolism:

A prime focus in pharmacogenomics is on genes encoding drug-metabolising enzymes, particularly those in the cytochrome P450 (CYP450) family. Variants in CYP2D6, CYP2C19, and CYP3A5 significantly influence drug pharmacokinetics. For instance, CYP2D6 poor metabolisers exhibit reduced clearance of codeine, leading to subtherapeutic analgesia, while ultra-rapid metabolisers face a risk of toxicity due to excessive conversion to morphine. Similarly, polymorphisms in CYP2C19 affect the bioactivation of clopidogrel, influencing cardiovascular outcomes. Other gene-drug interactions include TPMT variants impacting thiopurine toxicity and UGT1A1 polymorphisms affecting irinotecan metabolism, directly affecting dosing and drug selection[46, 47].

Pharmacogenomic Databases and Guidelines:

To facilitate clinical implementation, curated databases such as PharmGKB and CPIC (Clinical Pharmacogenetics Implementation Consortium) provide comprehensive resources linking genetic variants to pharmacological phenotypes. PharmGKB catalogues gene-drug associations and clinical annotations, while CPIC issues evidence-based guidelines on adjusting drug therapy based on genotype. These platforms are integral to translating genomic data into actionable prescribing decisions.

Clinical Implementation and Barriers:

Despite the scientific maturity of pharmacogenomics, clinical adoption remains inconsistent. Factors limiting implementation include:
Limited clinician awareness or training, which reduces confidence in interpreting pharmacogenomic data.
The lack of standardised electronic health record (EHR) integration hinders providing support for real-time clinical decision-making.
Population diversity gaps in pharmacogenomic studies result in incomplete or biased dosing recommendations.
Reimbursement and regulatory uncertainty affect test accessibility and sustainability.
Several institutions have successfully embedded pharmacogenomic testing into clinical workflows, particularly in oncology, psychiatry, and cardiology. Some health systems now offer Preemptive genotyping panels to personalise therapy across multiple drug classes[48–51].

As pharmacogenomics increasingly intersects with broader multi-omics efforts, especially in cancer and rare diseases, its clinical relevance and utility are poised to expand further.

Emerging Technologies Impacting Multi-Omics

Recent technological innovations are dramatically reshaping the multi-omics landscape, enhancing resolution, throughput, and interpretability in drug response studies.

Long-Read Sequencing:

Short-read next-generation sequencing (NGS) platforms have been instrumental in genomics, yet they struggle to resolve complex genomic regions and structural variants. Long-read sequencing technologies, such as those offered by PacBio and Oxford Nanopore, overcome these limitations by reading full-length DNA or RNA molecules. This facilitates accurate detection of gene fusions, repeat expansions, and transcript isoforms—elements often critical in cancer drug resistance and pharmacogenetics[52].

Spatial Transcriptomics:

Spatial omics technologies allow researchers to map gene expression within tissue architecture, preserving spatial context lost in dissociative methods. Platforms like 10x Genomics Visium and NanoString GeoMx enable spatially resolved transcriptomics and proteomics, helping researchers understand how cellular microenvironments, such as immune niches or hypoxic zones, influence drug response. This is particularly transformative in studying tumours, where spatial heterogeneity dictates therapeutic outcomes[53].

AI-Powered Integrative Platforms:

Integrating AI and machine learning with multi-omics analysis accelerates the discovery of predictive biomarkers and therapeutic targets. Platforms like DeepMO and MultiOmics AI employ neural networks and multi-view learning to synthesise complex datasets and generate interpretable predictions. These models are increasingly being applied in clinical trials to stratify patients and predict treatment efficacy[54].

Cloud and High-Performance Computing (HPC):

The massive scale of multi-omics data necessitates advanced computational infrastructure. Cloud computing platforms (e.g., Google Cloud, AWS, and Terra) and HPC clusters provide the scalability required for large-scale integrative analysis, including real-time processing of sequencing data and training of deep learning models. These resources democratise access to high-end analytics for researchers across institutions[55]. Collectively, these technologies are expanding the frontiers of multi-omics and bridging the translational gap between research insights and clinical applications.

Time-Series & Longitudinal Omics Analysis of Drug Response

Traditional omics studies offer static snapshots of biological states, limiting their ability to capture dynamic responses to therapy. Time-series and longitudinal multi-omics analyses provide a richer framework for understanding the temporal evolution of drug effects, resistance emergence, and treatment adaptation.

Temporal Profiling in Drug Response:

Temporal omics profiling involves sampling biological material at multiple time points during treatment, allowing researchers to monitor molecular changes over time. For example, tracking gene expression, protein phosphorylation, or metabolite levels can reveal early biomarkers of therapeutic efficacy or impending resistance. In immune-oncology, longitudinal transcriptomic profiling of tumour biopsies and circulating immune cells during checkpoint blockade therapy has identified early signatures predictive of response or immune-related adverse events[56].

Computational Models for Dynamic Data:

To analyse time-series omics, specialised computational models are needed:

Dynamic Bayesian Networks (DBNs): These probabilistic graphical models infer time-dependent regulatory interactions and are well-suited for modelling gene networks under drug perturbation.

Time-Series Clustering Algorithms: Methods such as k-means with dynamic time warping (DTW) or Gaussian process-based clustering can identify trajectories of molecular features across time points.

These models help identify transient regulators, switch-like gene behaviour, and tipping points that precede phenotypic changes[57].

Ethical, Legal, and Data Privacy Considerations

As multi-omics research expands in scope and depth, it raises significant ethical, legal, and privacy challenges. These issues are particularly acute due to the inherently identifiable nature of multi-layered biological data, which increases the risk of participant re-identification even when traditional de-identification techniques are applied.

Risks of Re-Identification:

Unlike isolated genomic data, multi-omics datasets, when combined with clinical metadata, create detailed molecular fingerprints of individuals. Even anonymised datasets can be vulnerable to linkage attacks when cross-referenced with publicly available information or environmental exposure records. The high resolution and longitudinal nature of omics profiling further amplify this risk, underscoring the need for robust privacy-preserving protocols[58].

Consent Frameworks and Data Governance:

Traditional informed consent models may be insufficient for multi-omics research, which often involves future, unspecified uses of data. Broad consent frameworks, tiered consent, and dynamic consent platforms are being explored to give participants greater control over data usage. These models align with the FAIR (Findable, Accessible, Interoperable, Reusable) data principles, aiming to maximise omics data's utility while maintaining participant autonomy and trust[59].

Compliance with Data Protection Regulations:

Regulatory frameworks such as the General Data Protection Regulation (GDPR) in the EU and the Health Insurance Portability and Accountability Act (HIPAA) in the U.S. establish strict guidelines for collecting, storing, and sharing personal health data. Multi-omics studies must navigate these regulations carefully, particularly in cross-border research collaborations. Compliance requires secure data storage, access control, audit trails, and clear data-sharing agreements. In addition, ethical review boards and data access committees play critical roles in evaluating omics-driven studies' societal implications and governance structures. Future policies must evolve to address the unique challenges of increasingly granular and longitudinal datasets, balancing innovation with respect for individual rights and privacy[60].

Future Directions & Unmet Needs

Despite remarkable progress, several gaps remain in integrating multi-omics into precision medicine. Addressing these challenges will require coordinated efforts across technological, methodological, and social domains.

Standardisation and Benchmarking:

One of the most pressing needs is the establishment of standardised protocols for data generation, preprocessing, and integration. Variability in sample processing, data normalisation, and feature selection undermines reproducibility and comparability across studies. Benchmark datasets and consensus pipelines similar to those in genomics (e.g., GATK) are urgently needed for multi-omics. Moreover, performance benchmarks for integration algorithms and predictive models are lacking. Community-driven challenges and open competitions like DREAM challenges could help establish gold standards and catalyse innovation[61].

Integration with Digital Health and Wearables:

Integrating with digital health technologies offers exciting possibilities as multi-omics moves toward clinical implementation. Wearables and remote sensors can provide real-time physiological data that complement molecular profiles, enabling continuous drug efficacy and toxicity monitoring. Coupling omics data with behavioural and environmental inputs from mobile health apps can further contextualise treatment response, paving the way for real-time, adaptive therapy[62].

Advances in Trans-Omics and Real-Time Analytics:

Beyond traditional multi-omics, the emerging field of trans-omics aims to capture direct causal relationships between omics layers, such as how transcription factors (proteins) regulate gene expression or how metabolite flux influences signalling networks. These models offer deeper mechanistic insight but require time-resolved and perturbation-based data to be effective. Real-time analytics, powered by edge computing and federated learning, could allow on-the-fly integration and interpretation of omics data without centralising sensitive datasets. This is particularly relevant for large, multi-centre clinical trials and decentralised care models[63].

Diversity and Inclusion in Data Collection:

A major limitation in current multi-omics research is the underrepresentation of diverse populations. Most reference datasets are skewed toward individuals of European descent, limiting the generalizability of predictive models. Expanding multi-omics efforts to include ethnically and geographically diverse cohorts is critical to ensuring that the benefits of precision medicine are equitably distributed. Efforts like the All of Us Research Program in the U.S. and the H3Africa initiative exemplify moves toward more inclusive omics research. Incorporating social determinants of health into multi-omics frameworks will also be essential for capturing the full spectrum of factors influencing drug response[64].

CONCLUSION

Integrating multi-omics technologies into drug response research marks a pivotal shift toward truly personalised medicine. Researchers can construct a comprehensive view of the biological processes influencing therapeutic outcomes by capturing diverse molecular layers genomics, transcriptomics, proteomics, metabolomics, and epigenomics. Throughout this review, we have highlighted how each omics domain contributes unique and synergistic insights into pharmacodynamics and pharmacokinetics. We have examined the computational strategies ranging from statistical integration models to AI-driven platforms that enable the fusion of complex datasets into actionable predictions. Real-world case studies and clinical trials demonstrate the tangible benefits of multi-omics in guiding treatment decisions. At the same time, emerging tools such as single-cell and spatial omics further illuminate drug response heterogeneity at unprecedented resolution. Nonetheless, significant challenges remain, including data heterogeneity, computational complexity, privacy concerns, and the need for greater diversity and standardisation. Addressing these issues will require multidisciplinary collaboration, robust infrastructure, and policy innovation. Looking forward, the future of precision drug therapy lies in the seamless integration of multi-omics with digital health, real-time analytics, and inclusive data practices. By advancing these frontiers, we move closer to a healthcare paradigm that is more precise, equitable, responsive, and effective.

Conflict of interest:

The authors declare no conflict of interest.

REFERENCE

Di Nicolantonio F, Vitiello PP, Marsoni S, et al (2021) Precision oncology in metastatic colorectal cancer — from biology to medicine. Nat Rev Clin Oncol 18:506–525. https://doi.org/10.1038/s41571-021-00495-z
Wörheide MA, Krumsiek J, Kastenmüller G, Arnold M (2021) Multi-omics integration in biomedical research – A metabolomics-centric review. Analytica Chimica Acta 1141:144–162. https://doi.org/10.1016/j.aca.2020.10.038
Periyasamy M (2025) AI-Driven Multi-Omics Integration for Enhanced Drug Discovery Pipelines. In: 2025 International Conference on Multi-Agent Systems for Collaborative Intelligence (ICMSCI). IEEE, Erode, India, pp 1553–1558
Madrid-Márquez L, Rubio-Escudero C, Pontes B, et al (2022) MOMIC: A Multi-Omics Pipeline for Data Analysis, Integration and Interpretation. Applied Sciences 12:3987. https://doi.org/10.3390/app12083987
Sánchez-Bayona R, Catalán C, Cobos MA, Bergamino M (2025) Pharmacogenomics in Solid Tumors: A Comprehensive Review of Genetic Variability and Its Clinical Implications. Cancers 17:913. https://doi.org/10.3390/cancers17060913
Malone ER, Oliva M, Sabatini PJB, et al (2020) Molecular profiling for precision cancer therapies. Genome Med 12:8. https://doi.org/10.1186/s13073-019-0703-1
Xicota L, De Toma I, Maffioletti E, et al (2022) Recommendations for pharmacotranscriptomic profiling of drug response in CNS disorders. European Neuropsychopharmacology 54:41–53. https://doi.org/10.1016/j.euroneuro.2021.10.005
Guevara-Nieto HM, Parra-Medina RS, Zabaleta J, et al (2024) Abstract 7609: Potential gene expression meta-signatures predict neoadjuvant chemotherapy response in invasive breast cancer. Cancer Research 84:7609–7609. https://doi.org/10.1158/1538-7445.AM2024-7609
Zou M, Zhou H, Gu L, et al (2024) Therapeutic Target Identification and Drug Discovery Driven by Chemical Proteomics. Biology 13:555. https://doi.org/10.3390/biology13080555
Knutson TP, Lange CA (2014) Tracking progesterone receptor-mediated actions in breast cancer. Pharmacology & Therapeutics 142:114–125. https://doi.org/10.1016/j.pharmthera.2013.11.010
Bowen TJ, Southam AD, Hall AR, et al (2023) Simultaneously discovering the fate and biochemical effects of pharmaceuticals through untargeted metabolomics. Nat Commun 14:4653. https://doi.org/10.1038/s41467-023-40333-7
Rawat PS, Jaiswal A, Khurana A, et al (2021) Doxorubicin-induced cardiotoxicity: An update on the molecular mechanism and novel therapeutic strategies for effective management. Biomedicine & Pharmacotherapy 139:111708. https://doi.org/10.1016/j.biopha.2021.111708
Yang L, Jin M, Jeong KW (2021) Histone H3K4 Methyltransferases as Targets for Drug-Resistant Cancers. Biology 10:581. https://doi.org/10.3390/biology10070581
Tao F, Zhang Z (2023) Editorial: Epigenetic drugs and therapeutic resistance for epithelial malignancies. Front Pharmacol 14:1208518. https://doi.org/10.3389/fphar.2023.1208518
Liu Y, Lian G, Chen T (2024) A novel multi-omics data analysis of dose-dependent and temporal changes in regulatory pathways due to chemical perturbation: a case study on caffeine. Toxicology Mechanisms and Methods 34:164–175. https://doi.org/10.1080/15376516.2023.2265462
Chai H, Zhou X, Zhang Z, et al (2021) Integrating multi-omics data through deep learning for accurate cancer prognosis prediction. Computers in Biology and Medicine 134:104481. https://doi.org/10.1016/j.compbiomed.2021.104481
Ba R, Geffard E, Douillard V, et al (2022) Surfing the Big Data Wave: Omics Data Challenges in Transplantation. Transplantation 106: e114–e125. https://doi.org/10.1097/TP.0000000000003992
Zhang Z, Zhang X (2024) Data-driven batch detection enhances single-cell omics data analysis. Cell Systems 15:893–894. https://doi.org/10.1016/j.cels.2024.09.011
Krassowski M, Das V, Sahu SK, Misra BB (2020) State of the Field in Multi-Omics Research: From Computational Needs to Data Mining and Sharing. Front Genet 11:610798. https://doi.org/10.3389/fgene.2020.610798
Ciceri S, Carenzo A, Iannó MF, et al (2022) Gene expression-based dissection of inter-histotypes, intra-histotype and intra-tumor heterogeneity in pediatric tumors. Sci Rep 12:17837. https://doi.org/10.1038/s41598-022-20536-6
Santorsola M, Lescai F (2023) The promise of explainable deep learning for omics data analysis: Adding new discovery tools to AI. New Biotechnology 77:1–11. https://doi.org/10.1016/j.nbt.2023.06.002
Subramanian I, Verma S, Kumar S, et al (2020) Multi-omics Data Integration, Interpretation, and Its Application. Bioinform Biol Insights 14:117793221989905. https://doi.org/10.1177/1177932219899051
Picard M, Scott-Boyer M-P, Bodein A, et al (2021) Integration strategies of multi-omics data for machine learning analysis. Computational and Structural Biotechnology Journal 19:3735–3746. https://doi.org/10.1016/j.csbj.2021.06.030
Pierre-Jean M, Deleuze J-F, Le Floch E, Mauger F (2020) Clustering and variable selection evaluation of 13 unsupervised methods for multi-omics data integration. Briefings in Bioinformatics 21:2011–2030. https://doi.org/10.1093/bib/bbz138
Argelaguet R, Arnol D, Bredikhin D, et al (2020) MOFA+: a statistical framework for comprehensive integration of multi-modal single-cell data. Genome Biol 21:111. https://doi.org/10.1186/s13059-020-02015-1
Demirel HC, Arici MK, Tuncbag N (2022) Computational approaches leveraging integrated connections of multi-omic data toward clinical applications. Mol Omics 18:7–18. https://doi.org/10.1039/D1MO00158B
Guarino V, Gliozzo J, Clarelli F, et al (2023) Intrinsic-Dimension Analysis for Guiding Dimensionality Reduction in Multi-Omics Data: In: Proceedings of the 16th International Joint Conference on Biomedical Engineering Systems and Technologies. SCITEPRESS - Science and Technology Publications, Lisbon, Portugal, pp 243–251
Pache MM, Pangavhane RR, Jagtap MN, Darekar AB (2025) The AI-Driven Future of Drug Discovery: Innovations, Applications, and Challenges. Asian J Res Pharm Sci 15:61–67. https://doi.org/10.52711/2231-5659.2025.00009
Gschwind A, Ossowski S (2025) AI Model for Predicting Anti-PD1 Response in Melanoma Using Multi-Omics Biomarkers. Cancers 17:714. https://doi.org/10.3390/cancers17050714
Sharma A, Lysenko A, Boroevich KA, Tsunoda T (2023) DeepInsight-3D architecture for anti-cancer drug response prediction with deep-learning on multi-omics. Sci Rep 13:2483. https://doi.org/10.1038/s41598-023-29644-3
Malik V, Kalakoti Y, Sundar D (2021) Deep learning assisted multi-omics integration for survival and drug-response prediction in breast cancer. BMC Genomics 22:214. https://doi.org/10.1186/s12864-021-07524-2
Rashid MM, Selvarajoo K (2024) Advancing drug-response prediction using multi-modal and -omics machine learning integration (MOMLIN): a case study on breast cancer clinical data. Briefings in Bioinformatics 25: bbae300. https://doi.org/10.1093/bib/bbae300
Mitra S, Saha S, Hasanuzzaman M (2020) Multi-view clustering for multi-omics data using unified embedding. Sci Rep 10:13654. https://doi.org/10.1038/s41598-020-70229-1
Degenhardt F, Seifert S, Szymczak S (2019) Evaluation of variable selection methods for random forests and omics data sets. Briefings in Bioinformatics 20:492–503. https://doi.org/10.1093/bib/bbx124
Wang Z, Li H, Carpenter C, Guan Y (2020) Challenge-Enabled Machine Learning to Drug-Response Prediction. AAPS J 22:106. https://doi.org/10.1208/s12248-020-00494-5
Petrelli F, Borgonovo K, Cabiddu M, Barni S (2012) Efficacy of EGFR Tyrosine Kinase Inhibitors in Patients With EGFR-Mutated Non–Small-Cell Lung Cancer: A Meta-Analysis of 13 Randomized Trials. Clinical Lung Cancer 13:107–114. https://doi.org/10.1016/j.cllc.2011.08.005
Lukashchuk N, Armenia J, Tobalina L, et al (2022) BRCA reversion mutations mediated by microhomology-mediated end joining (MMEJ) as a mechanism of resistance to PARP inhibitors in ovarian and breast cancer. JCO 40:5559–5559. https://doi.org/10.1200/JCO.2022.40.16_suppl.5559
Heo YJ, Hwa C, Lee G-H, et al (2021) Integrative Multi-Omics Approaches in Cancer Research: From Biological Networks to Clinical Subtypes. Molecules and Cells 44:433–443. https://doi.org/10.14348/molcells.2021.0042
Jariwala PB, Pellock SJ, Goldfarb D, et al (2020) Discovering the Microbial Enzymes Driving Drug Toxicity with Activity-Based Protein Profiling. ACS Chem Biol 15:217–225. https://doi.org/10.1021/acschembio.9b00788
M. Pache M, R. Pangavhane R (2025) Immunotherapy in Autoimmune Diseases: Current Advances and Future Directions. AJPR 183–191. https://doi.org/10.52711/2231-5691.2025.00030
Zimmermann M, Zimmermann-Kogadeeva M, Wegmann R, Goodman AL (2019) Separating host and microbiome contributions to drug pharmacokinetics and toxicity. Science 363: eaat9931. https://doi.org/10.1126/science.aat9931
Cheng X, Wang J, Gong L, et al (2022) Composition of the Gut Microbiota Associated with the Response to Immunotherapy in Advanced Cancer Patients: A Chinese Real-World Pilot Study. JCM 11:5479. https://doi.org/10.3390/jcm11185479
Yan G, Song D, Li JJ (2023) scReadSim: a single-cell RNA-seq and ATAC-seq read simulator. Nat Commun 14:7482. https://doi.org/10.1038/s41467-023-43162-w
Quek C, Pratapa A, Bai X, et al (2024) Single-cell spatial multiomics reveals tumor microenvironment vulnerabilities in cancer resistance to immunotherapy. Cell Reports 43:114392. https://doi.org/10.1016/j.celrep.2024.114392
Pache M, Nikam S (2025) Antibiotic Resistance: Current Challenges and Future Directions. Int J of Pharm Sci 3:1600–1622. https://doi.org/10.5281/ZENODO.14690670
Milosavljevi? F, Leucht S (2024) Living systematic review and meta-analysis of plasma-concentrations of antipsychotic drugs in carriers and non-carriers of variant CYP450 genotypes: Living systematic review protocol. F1000Res 13:452. https://doi.org/10.12688/f1000research.147794.2
Pache MM, Pangavhane RR, Nikam SV, et al (2025) CRISPR-Cas9 in Pharmaceutical Research: Applications, Challenges, Ethical Considerations and Future Directions. RJST 296–304. https://doi.org/10.52711/2231-5713.2025.00045
Kabbani D, Akika R, Wahid A, et al (2023) Pharmacogenomics in practice: a review and implementation guide. Front Pharmacol 14:1189976. https://doi.org/10.3389/fphar.2023.1189976
Relling MV, Klein TE (2011) CPIC: Clinical Pharmacogenetics Implementation Consortium of the Pharmacogenomics Research Network. Clin Pharmacol Ther 89:464–467. https://doi.org/10.1038/clpt.2010.279
Gaedigk A, Whirl?Carrillo M, Pratt VM, et al (2020) PharmVar and the Landscape of Pharmacogenetic Resources. Clin Pharma and Therapeutics 107:43–46. https://doi.org/10.1002/cpt.1654
Pache M, Kedar H, Kond S, Jadhav P (2025) Pharmacological Management of Neurodegenerative Disorders: Current and Future Approaches. Int J Sci R Tech 2:405–520. https://doi.org/10.5281/ZENODO.15074000
Eisfeldt J, Higginbotham EJ, Lenner F, et al (2024) Resolving complex duplication variants in autism spectrum disorder using long-read genome sequencing. Genome Res 34:1763–1773. https://doi.org/10.1101/gr.279263.124
Che G, Yin J, Wang W, et al (2024) Circumventing drug resistance in gastric cancer: A spatial multi-omics exploration of chemo and immuno-therapeutic response dynamics. Drug Resistance Updates 74:101080. https://doi.org/10.1016/j.drup.2024.101080
Zafari N, Bathaei P, Velayati M, et al (2023) Integrated analysis of multi-omics data for the discovery of biomarkers and therapeutic targets for colorectal cancer. Computers in Biology and Medicine 155:106639. https://doi.org/10.1016/j.compbiomed.2023.106639
Kaur P, Singh A, Chana I (2021) Computational Techniques and Tools for Omics Data Analysis: State-of-the-Art, Challenges, and Future Directions. Arch Computat Methods Eng 28:4595–4631. https://doi.org/10.1007/s11831-021-09547-0
Levy E, Milani P, Navarro F, et al (2021) 75 Comprehensive profiling of the tumor-immune microenvironment using an augmented transcriptome. J Immunother Cancer 9: A83–A83. https://doi.org/10.1136/jitc-2021-SITC2021.075
Ruiz-Perez D, Lugo-Martinez J, Bourguignon N, et al (2021) Dynamic Bayesian Networks for Integrating Multi-omics Time Series Microbiome Data. mSystems 6:10.1128/msystems.01105-20. https://doi.org/10.1128/msystems.01105-20
Dupras C, Bunnik EM (2021) Toward a Framework for Assessing Privacy Risks in Multi-Omic Research and Databases. The American Journal of Bioethics 21:46–64. https://doi.org/10.1080/15265161.2020.1863516
Wiertz S, Boldt J (2022) Evaluating models of consent in changing health research environments. Med Health Care and Philos 25:269–280. https://doi.org/10.1007/s11019-022-10074-3
Mohr AE, Ortega-Santos CP, Whisner CM, et al (2024) Navigating Challenges and Opportunities in Multi-Omics Integration for Personalized Healthcare. Biomedicines 12:1496. https://doi.org/10.3390/biomedicines12071496
Ewald JD, Zhou G, Lu Y, et al (2024) Web-based multi-omics integration using the Analyst software suite. Nat Protoc 19:1467–1497. https://doi.org/10.1038/s41596-023-00950-4
Tong L, Shi W, Isgut M, et al (2024) Integrating Multi-Omics Data With EHR for Precision Medicine Using Advanced Artificial Intelligence. IEEE Rev Biomed Eng 17:80–97. https://doi.org/10.1109/RBME.2023.3324264
Newman NK, Macovsky MS, Rodrigues RR, et al (2024) Transkingdom Network Analysis (TkNA): a systems framework for inferring causal factors underlying host–microbiota and other multi-omic interactions. Nat Protoc 19:1750–1778. https://doi.org/10.1038/s41596-024-00960-w
Yang G, Mishra M, Perera MA (2023) Multi?Omics Studies in Historically Excluded Populations: The Road to Equity. Clin Pharma and Therapeutics 113:541–556. https://doi.org/10.1002/cpt.2818.