We've updated our Privacy Policy to make it clearer how we use your personal data. We use cookies to provide you with a better experience. You can read our Cookie Policy here.


Multiomics – A Multi-Layered Answer to Multi-Layered Questions

A graphical recreation of a human hologram from the shoulders upwards.
Credit: iStock
Listen with
Register for free to listen to this article
Thank you. Listen to this article using the player above.

Want to listen to this article for FREE?

Complete the form below to unlock access to ALL audio articles.

Read time: 8 minutes

Omics are untargeted technologies that simultaneously profile all molecules of a specific type in a biological system. Multiomics, or integrative omics, has emerged as a method to integrate two or more omics to generate comprehensive profiles of biological systems. A PubMed search of “multiomics” yields 7,463 results*, of which 5,539 results or 74%, were published in just the last three years. The approach can predict biological characteristics or generate granular multilevel information on biological systems. Basic research applications span cancer biology, disease pathophysiology, tissue dynamics and host-virus interactions, among many others. Multiomics also has clinical applications, such as drug and biomarker discovery, personalized medicine, and clinical trial evaluation. More recently, the multiomics approach is being leveraged at the single-cell level to generate granular information on heterogeneity in cell populations compared to bulk methods. Single-cell methods also increases the chance of yielding mechanistic insight, as co-expression and interaction can be tied to individual cells.

As a novel approach, new research and clinical applications of multiomics continue to emerge,  tackling complex topics.

Single Cell Buyer’s Guide

Cell heterogeneity is a key contributor to biological complexity that is often masked by bulk techniques, like RNA sequencing or microarray analysis. Rather than providing an average snapshot of how cells work, single cell sequencing technology gives researchers the ability to characterize tissue heterogeneity in more detail, identify rare cell types and dissect molecular mechanisms cell by cell. Download this guide to explore single cell gene expression, immune profiling and more.

Download Guide

Glorious glycans! Multiomics sheds light

Glycans are linear or branched polymers of various monosaccharides, which differentiates them from most other biomolecules, such as nucleic acids and proteins, that are generally formed from linearly joined building blocks. “The ability to branch vastly expands the complexity of glycans and challenges analyses of their interactions with other biomolecules,” explained Daniel Bojar, professor at the Wallenberg Centre for Molecular and Translational Medicine at the University of Gothenburg. “Whereas glycans might appear similar written out on paper, the branching pattern results in structural characteristics that dictate biological properties. And although the field has a tendency to treat protein glycans, lipid glycans, etc., as distinct entities, we are now shifting to a view that considers the cumulative distribution of all glycans in a living cell, known as the glycome.”

“I think the glycome has gone relatively underappreciated compared to other ‘omes’,” Bojar continued. “This is not to say that glycans are not biologically important. On the contrary, glycosylated biomolecules regulate many important molecular processes, such as protein folding and ligand-receptor interactions. Glycans are also involved in diseases, such as immune disorders, cancer and infectious disease through host-pathogen interactions.” Bojar and his team are developing glycoinformatics tools, which are computational methods that connect the dots in the glycome to shed insight into glycan structure and function. Among the tools are Glycowork, an open-source software platform to analyze glycans, SweetNet, a neural network model to predict glycan properties and LectinOracle, a deep learning model to predict lectin–glycan binding.

In addition to the complexity incurred by their branching structure, glycan biosynthesis is also complicated. Glycans are not generated by a templated process, like proteins from nucleic acids. During translation, each three-base codon in the mRNA molecule specifies a particular amino acid in the protein sequence. “Glycans are generated by a biosynthetic network comprised of hundreds of enzymes and associated proteins,” Bojar elaborated.

Glycosylation biosynthetic enzymes, which are encoded by so-called “glycogenes,” are transcriptionally and translationally regulated, rendering the glycome a highly dynamic entity dependent on the cellular state. “The glycome can be affected by genetic, transcriptomic, proteomic and metabolic changes. Since several biological processes contribute to the distribution of glycans in the cell, multiomics can be a very valuable tool to study how the cellular state influences the glycome,” Bojar explained.

With this goal in mind, Bojar and his team launched a single-cell, multiomics analysis of mouse T cells, in collaboration with the group of Dr. Lara K. Mahal at the University of Alberta. The study analysis leveraged a dataset of single-cell measurements called SUrface-protein Glycan And RNA-seq (SUGAR-seq), which simultaneously profiles the cell’s transcriptome by single-cell RNA-seq and surface β1,6-branched glycans using a lectin binding probe.

A Storm We Can’t Weather… Yet.

In a normal immune response, entry of the SARS-CoV-2 virus into lung tissue induces controlled activation of the immune system, however, cytokine release syndrome can occur if there is an imbalance between pro- and anti-inflammatory cytokines and chemokines. Download this poster to explore the role of immune cells and effector molecules in cytokine release syndrome.

Download Poster

“SUGAR-seq is a really exciting method,” Bojar said of the platform. “There were no single-cell methods before it to simultaneously analyze the cell’s transcriptome and aspects of the glycome. Even bulk analysis of the glycome faces issues. So, the single-cell SUGAR-seq platform really unlocked the possibility of leveraging multiomics analysis to identify mechanistic links between the cellular transcriptome and its surface glycome.”

Bojar and his team applied deep learning to the SUGAR-seq database of mouse tumor-infiltrating T cells to develop a model that could predict the surface glycome from the transcriptome. Next, they employed a method called SHapley Additive exPlanations (SHAP)  to identify gene transcripts that correlated with low or high β1,6-branched glycan abundance. “We then ran pathway enrichment analyses of SHAP genes, which pinpointed processes that negatively regulate T cell activity and differentiation. SHAP genes revolved around the biology of β1,6-branched glycan and MGAT5, the enzyme that forges the β1,6-branch of N-glycans,” Bojar elaborated of the study findings. “Additional important genes were also identified, especially related to cytokine receptor interactions and immune suppression.”

Overall, the study demonstrated how single-cell multiomics could be leveraged to identify gene expression patterns that disentangled the multilayered roles of surface glycosylation, from biosynthesis to functions in immunomodulation.

Bojar thinks that multiomics in glycobiology is just warming up and that there is a long road ahead. “I think cancer is on the research frontier. Sialic acid glycans are upregulated in cancer and help the tumor evade the immune system and metastasize, though in different structural configurations. Multiomics analyses have been conducted in melanoma to pinpoint candidate genes that promote aberrant glycans, and further studies may suggest possible therapeutic avenues that exploit glycan-mediated processes,” Bojar shared. “Another technical advancement that would propel the field forward would be expanding the toolkit of probes for profiling various glycan structural epitopes.”

A Fully Integrated Spatial Biology Analysis Solution 

Download this app note to discover a platform that can allow you to visualize and store spatial multiomic data on a single platform, share data seamlessly and collaborate with researchers at your institution or across the globe and analyze data easily.

View App Note

Multiomics makes medical discovery: Peripheral neuropathy

In addition to advances in basic research, multiomics offers vast potential in biomedical research. “Most diseases are highly complex and progress through a concerted breakdown in multiple biological processes,” explained Junguk Hur, professor in the Department of Biomedical Sciences at the University of North Dakota. “To truly gain insight into disease pathophysiology, it is essential to examine several levels of biological systems, and multiomics is an excellent tool to facilitate that. In addition to gaining insight into disease processes, it is also possible to identify potential therapeutics through drug repurposing by integrating omics and multiomics with the drug structure space and drug-gene perturbation.”

Toward these goals, Hur and his team develop bioinformatics tools to deepen understanding of disease pathophysiology and discover potential drug candidates on a spectrum of illnesses.

“An important interest in our laboratory is diabetic peripheral neuropathy (DPN), which damages the peripheral nerves in patients with diabetes. DPN is primarily caused by metabolic, i.e., environmental, stresses, and epigenetics is increasingly recognized as a contributing factor. Thus, DPN is a prime candidate for applying multiomics,” Hur explained.

In a study of nerve biopsies from patients with well versus poorly controlled diabetes, Hur and his team, in collaboration with Dr. Eva Feldman at Michigan Medicine, integrated nerve transcriptomics with genome-wide DNA methylation. The resulting functional and network analysis suggested that genes and pathways regulating immune response, extracellular matrix remodeling and cell cycle most strongly differentiated the neuropathic nerve in patients with well versus poorly controlled diabetes. “This was an exciting finding because it implicated, for the first time, that DNA methylation may regulate gene expression in DPN, which we were able to perceive using multiomics. It also suggested that epigenetic regulation could be linked to glycemic control in patients,” Hur concluded.

In a further preclinical study of DPN in diabetic mice, Hur and Feldman groups conducted a multiomics lipidomic and transcriptomic evaluation of the neuropathic nerve. The resultant lipid–gene transcript network highlighted essential genes related to lipid metabolism and transport, including acyltransferase 2 (Dgat2). “We confirmed the relevance of Dgat2 to DPN in nerve biopsies from human patients with diabetes and found that elevated nerve DGAT2 protein correlated with hyperlipidemia,” Hur elaborated. DGAT2 catalyzes the final and only committed step in triglyceride synthesis and is linked to injury in metabolically active tissues. Inhibiting DGAT2 promotes nerve regeneration, underscoring its importance to DPN. “Therefore, our study illustrated how multiomics can be applied to generate biological insight and pinpoint candidates for further study,” Hur summarized of the investigation.

An Integrated Multiomic Approach for the Diagnosis of Metabolic Diseases

Early disease recognition and treatment are, in many cases, critical for preventing neurological impairment and/or death of the patient. Download this whitepaper to discover a panel that allows the efficient diagnosis of more than 180 metabolic diseases and obtained a diagnostic yield comparable or even higher than the average yield after exome sequencing.

Download Whitepaper

Multiomics for inflammatory disease

Inflammation is another prime candidate for multiomics analysis, due to the complex pathophysiology arising from host interactions with infectious agents or a dysregulayed microbiome.

“Another major focus for us has been to seek possible therapeutic solutions for inflammation, which is a frequent cause and consequence of disease. We recently collaborated on a multiomics project on inflammatory bowel disease (IBD), a complex chronic inflammation of the gastrointestinal tract, which progresses through host-microbiome interactions, especially via host innate and adaptive immunity,” Hur described.

The study used data from the IBD multiomics database, a longitudinal multi-tissue analysis of the host transcriptome and metagenome (i.e., the genomic analysis of the bacteria present in the host microbiome). “The host transcriptome–metagenome functional enrichment network, i.e., the network built from the functional role of the most highly correlated host-to-metagenome transcripts, identified cGAS-STING as an IBD biomarker.” cGAS-STING is an arm of the innate immune system, which launches a pro-inflammatory response upon sensing cytosolic DNA, presumably from bacterial infection. To validate cGAS-STING as a possible therapeutic target, the investigators generated cGAS knockout mice, which were more resistant to IBD induction.

“Once we had identified cGAS-STING as a potential target through multiomics, we leveraged the Library of Integrated Network-Based Cellular Signatures (LINCS), a drug perturbation database, to pinpoint possible therapeutics for IBD based on transcriptomic differences between the wild-type and cGAS knockout mice,” Hur elaborated on the next steps of the analysis. LINCS suggested 43 drug candidates that could perturb the disease cGAS-STING signature. “We focused on brefeldin A, an inhibitor of protein transport, which improved IBD symptoms and attenuated colon inflammation in mice. Overall, our multiomics analysis demonstrated how we could identify IBD biomarkers that may be amenable to treatment,” summarized Hur.

Hur and his group, in collaboration with other research laboratories, are also investigating adverse drug reactions, drug repurposing for viral infection and personalized medicine platforms using multiomics and integrative approaches.

Multiomics is a useful tool to uncover biological mechanisms and disease pathophysiology. It is especially important for understanding complex processes that occur through the interaction of multiple omics biology. Single-cell platforms further shed mechanistic insight by linking co-expression and interaction to individual cells. “The scope of multiomics applications is extensive, and we are just now tapping into the full potential for a range of diseases,” Hur concluded. “Moreover, new methodology of analyzing multiomics data is further widening the available toolkit.”

*as of October 2022.