We comprehensively reviewed the current medications and investigational drugs for COVID-19 using UpToDate and DynaMed databases. We identified 38 drugs or compounds as follows: glucocorticoids, including dexamethasone, hydrocortisone, and methylprednisolone; anti-viral agents, such as remdesivir, favipiravir, and ritonavir; immunomodulators including hydroxychloroquine and chloroquine; JAK inhibitors such as baricitinib; antiparasitic drugs including ivermectin; and other compounds under investigation, such as azithromycin, fluvoxamine, and famotidine (Table 1). Of these 38 drugs, 25 drugs could retrieve their corresponding target genes (Figure 1A) from CLUE database. A total of 30 target genes was obtained (Figure 1B; Table 1). We also sorted the inflammatory signaling pathways in COVID-19-related ARDS, including those of IL-6/JAK/STAT, interferon, NF-κB, TLRs, Bruton’s tyrosine kinase, and renin–angiotensin system (Zhang and Guo, 2020; Choudhary et al., 2021), and 49 target genes were included for further analysis (Figure 1C). The union of 30 and 49 target genes were used to searching for their corresponding TCM drugs via SymMap (Supplementary Tables S1, S2; filtered by p < 0.05). Each target gene may connect to different number of TCM drugs. For example, TNF, a highly frequent gene appearing in SymMap, connected up to 441 TCM drugs, whereas TLR3 only linked to 1 TCM. We then selected 10 highly frequent target genes (each connection with > 200 TCMs for each gene) and intersected their TCM sets to obtain 66 TCM candidates, which might potentially inhibit ARDS and COVID-19-related inflammatory response (Figures 1D,E; Table 2). Therefore, every single TCM listed in Table 2 links to 10 highly frequent target genes, including TNF, CYP1A2, CYP3A4, NOS2, IL6, IL1B, STAT1, NFKBIA, CXCL10, and IFNG. These 66 TCM candidates included a broad spectrum of common therapeutic classes, including antipyretics, antitussives, antiasthmatics, and Qi-reinforcing drugs. However, most of the predicted TCM drugs have not been thoroughly tested their inhibitory activities against SARS-CoV-2 (Lee et al., 2021; Jia et al., 2021). To prioritize potential candidates from these 66 TCM candidates, the following three criteria were considered: Firstly, we hypothesized that in silico identified potential anti-SARS-CoV-2 TCMs covering a wide range of pharmacologic functions with minimal side effects (Table 2) could be integrated into our clinical practice. Secondly, two reports utilized statistical calculation of the frequently used TCMs for SARS-CoV-2 infection in China (Luo et al., 2020; Xu et al., 2020). They identified 19 frequently used TCM in COVID-19 treatment. Among them, Lonicera Japonica (honeysuckle) and Astragalus membranaceus (Huangqi) were overlapped with our 66 TCM list. Despite some TCMs are reported to contain nephrotoxins and mutagens (Ng et al., 2017), the toxicology of most of TCMs remains to be determined (Zeng and Jiang, 2010). Instead, Honeysuckle and Huangqi showed safety without distinct toxicity or side effects in various studies (Shang et al., 2011; Fu et al., 2014). Finally, our TCM combination includes the ingredients of heat-toxin clearing (honeysuckle) and qi-tonifying (Huangqi) comparing to other clinical trial medicines, which mainly consist of heat-toxin clearing agents and ignore the value of TCM in providing a supportive role in the treatment of COVID-19. Hence, we selected two low toxicities TCM drugs, honeysuckle and Huangqi, to further evaluate their anti-viral activities in silico and in vitro.203Please respect copyright.PENANAX0tZcn3oFz