Multiomics integration-based molecular characterizations of COVID-19

Abstract

The coronavirus disease 2019 (COVID-19) pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), rapidly became a global health challenge, leading to unprecedented social and economic consequences. The mechanisms behind the pathogenesis of SARS-CoV-2 are both unique and complex. Omics-scale studies are emerging rapidly and offer a tremendous potential to unravel the puzzle of SARS-CoV-2 pathobiology, as well as moving forward with diagnostics, potential drug targets, risk stratification, therapeutic responses, vaccine development and therapeutic innovation. This review summarizes various aspects of understanding multiomics integration-based molecular characterizations of COVID-19, which to date include the integration of transcriptomics, proteomics, genomics, lipidomics, immunomics and metabolomics to explore virus targets and developing suitable therapeutic solutions through systems biology tools. Furthermore, this review also covers an abridgment of omics investigations related to disease pathogenesis and virulence, the role of host genetic variation and a broad array of immune and inflammatory phenotypes contributing to understanding COVID-19 traits. Insights into this review, which combines existing strategies and multiomics integration profiling, may help further advance our knowledge of COVID-19.

Keywords: COVID-19; molecular characteristics; multiomics integration; outcome; severity; single-cell omics.

Figure 1

Applications of the multiomics integration-based molecular characterization of COVID-19. COVID-19, coronavirus disease 2019; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; MERS-CoV, Middle East respiratory syndrome CoV; ICU, intensive care unit; PACS, post-acute COVID syndrome. Created using BioRender.com.

Figure 2

Summary of five categories of multiomics integration strategies and their application in the molecular characterization of COVID-19. Commonly used multiomics data (left box) are integrated through five categories of integration approaches (middle) to investigate four major applications in the molecular characterization of COVID-19 (right box). The grey lines from the middle to the right represent the major applications of approaches for specific purposes. Both network-based and multistaged strategies have been performed for all four applications. Created using BioRender.com.

Figure 3

Illustration of the similarity-based assumptions for the transfer of previous knowledge to multiomics integration in COVID-19. Multiomics integration from immunomics, secretome, proteome, interactome, transcriptome, metabolome and lipidome amongst others could provide a systematic understanding of viral infection and COVID-19 disease progression and processes. SARS-CoV-2 is similar to SARS and MERS, as well as other viruses. Based on their similarities, the virus–host response and potential diagnostic and therapeutic targets derived from multiomics analyses could be transferred to and prioritized within COVID-19 research. Based on diseasome, given the similarity with known diseases, the candidate genes from diseases similar to COVID-19 could be analyzed and examined, particularly in relation to a predisposition to disease, comorbidities and in predicting long-COVID and characterizing patients. Drug similarities in terms of the chemical effects as well in the multiomics-level response could be used to prioritize candidate drugs and therapeutic targets. Created using BioRender.com.

Figure 4

Summary of platform co-appearance in 32 multiomics studies of COVID-19. Pie charts of the prevalence of various omics platforms (A) and biospecimen types (B) that appeared in at least two publications, respectively. C) A network plot of the co-appearance of omics platforms, in which omics (node) appeared in at least two publications and omics pairs (edge) co-appeared in at least two publications. The size of the nodes corresponds to their appearance number in these publications (3 to 21). The width of the edge is related to the number of the co-appearances in these publications (2 to 16). Detailed information is available in Supplementary Tables 2–4. PBMCs, peripheral blood mononuclear cells; BALF, bronchoalveolar lavage fluid. Pie charts and network are created using R version 3.6.0. and Cytoscape version 3.8.2, respectively.