Multi-omic analysis characterizes molecular susceptibility of receptors to SARS-CoV-2 spike protein

Figures

• 

  Fig. 1. The binding affinity analysis between human receptors and SARS-CoV-2 variant strains. (A) The optimal structural conformations of the spike proteins for the seven SARS-CoV-2 variants, as well as the best binding conformations between the BSG receptor and the spike protein chains of the variant strains, are as follows. Each variant's spike protein consists of three chains. The trimeric spike protein structures of the wild type, alpha, and beta variants were obtained from crystallized proteins in the PDB database, with the three monomers adopting a mixed conformation (one up, two down). The spike protein structures of the remaining four variants were obtained through homology modelling, and the monomers also adopted the same mixed conformation. In the binding conformation of omicron-BSG, the BSG receptor is represented in yellow, one chain of the omicron variant spike protein is represented in light blue, and the binding site is indicated by a red rectangle with dashed lines. (B) The heatmap shows the binding affinity. Darker red color indicates stronger affinity. (C) The lines show the trends of affinity changes for the receptors against different variant strains.

• 

  Fig. 2. The transcriptomic analysis at tissue level. (A) The bubble diagram illustrates the transcriptomic expression levels of 6 receptors across 22 human tissues/organs. (B) Comparative analysis of transcriptomic expression among different age groups. (C) Comparative analysis of transcriptomic expression among different underlying disease states (CVD, MD, and lung cancer).

• 

  Fig. 3. The transcriptomic expression in primates and the proteomics expression in human. (A) The transcriptomic expression heatmap displays the expression levels of various receptors in different tissues/organs of three primate animals, represented by CPM values. (B) The transcriptomic expression heatmap illustrates the expression levels of different receptors in various tissues/organs of three primate animals, using TPM values. (C) The proteomic expression heatmap showcases the expression of different receptors in human tissues/organs.

• 

  Fig. 4. The transcriptomic analysis for healthy individuals at single-cell level. The rows correspond to various tissues/organs, while the first column indicates the cell types present in those tissues/organs. The remaining columns represent the expression levels of each receptor in different tissues/organs. The scale has been normalized using log CPM.

• 

  Fig. 5. The single cell atlases for heart, liver, lung, and testis of healthy human, and specific receptors’ expression patterns on the atlases.

• 

  Fig. 6. The transcriptomic comparison between healthy and patients at single-cell level. The rows represent different tissues/organs in healthy or patient samples, while the first column displays the cell types corresponding to these tissues/organs. The remaining columns indicate the expression levels of each receptor in the different tissues/organs of healthy or patient samples. The expression levels have been normalized using log CPM.

• 

  Fig. 7. The single cell atlases for heart, liver, lung, and nasal mucosa between COVID-19 patients and healthy controls, and specific receptors’ expression patterns on the atlases.

• 

  Fig. 8. Cell types with SARS-CoV-2 RNA detected and specific receptors’ expression patterns in SARS-CoV-2 RNA-positive cells. (A) 13,916 SARS-CoV-2-RNA-positive cells were detected (UMI > 0) in the datasets of GSE145926 and GSE155249. (B) UMAP of the cell types of SARS-CoV-2 RNA-positive cells. (C) The marker genes used to determine cell types. (D) Viral load in SARS-CoV-2 RNA-positive cells quantified by log(CPM). (E) Expression levels of specific receptors in SARS-CoV-2-RNA-positive cells. (F) Pearson’s correlations of BSG, ASGR1 and TFRC’s expression patterns in different cell types (R>0.3, n > 500, p < 0.05) with viral load (CPM) (zero-expression cells were excluded from regression analysis to reduce the effects of dropouts).