Computational workflow for functional characterization of COVID-19 through secondary data analysis

Abstract

Summary Standard transcriptomic analyses cannot fully capture the molecular mechanisms underlying disease pathophysiology and outcomes. We present a computational heterogeneous data integration and mining protocol that combines transcriptional signatures from multiple model systems, protein-protein interactions, single-cell RNA-seq markers, and phenotype-genotype associations to identify functional feature complexes. These feature modules represent a higher order multifeatured machines collectively working toward common pathophysiological goals. We apply this protocol for functional characterization of COVID-19, but it could be applied to many other diseases. For complete details on the use and execution of this protocol, please refer to Ghandikota et al. (2021) .

The current status of gene expression profilings in COVID‐19 patients

“…There are some studies already published that performed secondary analysis of previously generated RNA‐seq and microarray data of COVID‐19 patients compared to healthy donors and individuals with other illnesses [e.g., SARS and the Middle East respiratory syndrome (MERS), lupus]. 26 , 27 , 28 , 29 Yet, to understand the disease mechanism of SARS‐CoV‐2, RNA‐seq data from long‐COVID patients should be generated not only from blood or blood‐related materials but also from tissue biopsy samples from the affected areas by SARS‐CoV‐2. Furthermore, more systematic analysis of RNA‐seq data combined with other OMICS data (e.g., genomics, proteomics, metabolomics), especially those of time‐course data, are urgently needed.…”

“…Such newly generated data can be compared to the previously generated data as listed in Table 1 to perform a comparative analysis of transcriptomic data to understand how gene expression changes affect COVID‐19 patients. There are some studies already published that performed secondary analysis of previously generated RNA‐seq and microarray data of COVID‐19 patients compared to healthy donors and individuals with other illnesses [e.g., SARS and the Middle East respiratory syndrome (MERS), lupus] 26–29 . Yet, to understand the disease mechanism of SARS‐CoV‐2, RNA‐seq data from long‐COVID patients should be generated not only from blood or blood‐related materials but also from tissue biopsy samples from the affected areas by SARS‐CoV‐2.…”