Network proximity measure
We posit that the human PPIs provide an unbiased, rational roadmap for repurposing drugs for potential treatment of HCoVs in which they were not originally approved. Given C, the set of host genes associated with a specific HCoV, and T, the set of drug targets, we computed the network proximity of C with the target set T of each drug using the “closest” method:1 \documentclass[12pt]{minimal} 				\usepackage{amsmath} 				\usepackage{wasysym} 				\usepackage{amsfonts} 				\usepackage{amssymb} 				\usepackage{amsbsy} 				\usepackage{mathrsfs} 				\usepackage{upgreek} 				\setlength{\oddsidemargin}{-69pt} 				\begin{document}$$\left\langle d_{CT} \right\rangle = \frac{1}{\vert\vert C \vert\vert +\vert\vert T \vert\vert}\left({\sum\limits_{c \in C}} {\min}_{t \in T}\,d\left( {c,t} \right) + {\sum\limits_{t \in T}} {\min}_{c \in C}\,d\left(c,t\right) \right),$$\end{document}dCT=1∣∣C∣∣+∣∣T∣∣∑c∈Cmint∈Tdc,t+∑t∈Tminc∈Cdc,t,where d(c, t) is the shortest distance between gene c and t in the human protein interactome. The network proximity was converted to Z-score based on permutation tests:2 \documentclass[12pt]{minimal} 				\usepackage{amsmath} 				\usepackage{wasysym} 				\usepackage{amsfonts} 				\usepackage{amssymb} 				\usepackage{amsbsy} 				\usepackage{mathrsfs} 				\usepackage{upgreek} 				\setlength{\oddsidemargin}{-69pt} 				\begin{document}$$Z_{d_{CT}} = \frac{{d_{CT} - \overline {d_r} }}{{\sigma _r}},$$\end{document}ZdCT=dCT−dr¯σr,where \documentclass[12pt]{minimal} 				\usepackage{amsmath} 				\usepackage{wasysym} 				\usepackage{amsfonts} 				\usepackage{amssymb} 				\usepackage{amsbsy} 				\usepackage{mathrsfs} 				\usepackage{upgreek} 				\setlength{\oddsidemargin}{-69pt} 				\begin{document}$$\overline {d_r}$$\end{document}dr¯ and σr were the mean and standard deviation of the permutation test repeated 1000 times, each time with two randomly selected gene lists with similar degree distributions to those of C and T. The corresponding P value was calculated based on the permutation test results. Z-score < −1.5 and P < 0.05 were considered significantly proximal drug–HCoV associations. All networks were visualized using Gephi 0.9.2 (https://gephi.org/).