Machine Learning and Network Medicine approaches for Drug Repositioning for COVID-19. Santos et al.

Supplemental Items used in "Machine Learning and Network Medicine approaches for Drug Repositioning for COVID-19" by Suzana
de Siqueira Santos, Mateo Torres, Diego Galeano, María del Mar Sánchez, Luca Cernuzzi, and Alberto Paccanaro. The items include
input datasets, and predictions by our matrix decomposition model and network medicine approach.

Predictions by our matrix decomposition model are shown in Supplementary File 1. Two out of three interactomes used by our
network medicine approach were mapped to SwissProt proteins and shown in Supplementary Files
2 (Luck et al., 2020, DOI: 10.1038/s41586-020-2188-x), and 3 (Cheng et al., 2018, DOI: 10.1038/s41467-018-05116-5).
For these two interactomes, we used drug-target associations in Supplementary File 4, and 336 host proteins
(UniProt accession numbers) in Supplementary File 5. For the Gysi el al. interactome
(Gysi et al., 2021, DOI: 10.1073/pnas.2025581118), we used Entrez IDs of the 336 host proteins shown in Supplementary File 6.
Supplementary File 7 shows the ATC categories of DrugBank drugs. DrugBank entries with effect against SARS-CoV-2 according to
CMAP, in vitro, and clinical trials evidence are shown in Supplementary Files 8, 9, and 10, respectively. Predictions by our network
medicine approach are shown in Supplementary File 11. The weights assigned to the host proteins by the kernel-based methods are
shown in Supplementary File 12.

本补充材料对应Suzana de Siqueira Santos、Mateo Torres、Diego Galeano、María del Mar Sánchez、Luca Cernuzzi及Alberto Paccanaro发表的题为《面向新型冠状病毒肺炎药物重定位的机器学习与网络医学方法》的研究论文，包含输入数据集以及本研究矩阵分解模型与网络医学方法生成的预测结果。本研究矩阵分解模型的预测结果见于补充文件1。本研究网络医学方法所使用的3个相互作用组中，有2个已映射至SwissProt蛋白（Swiss-Prot），相关数据分别收录于补充文件2（Luck等，2020，DOI: 10.1038/s41586-020-2188-x）与补充文件3（Cheng等，2018，DOI: 10.1038/s41467-018-05116-5）。针对上述两个相互作用组，本研究使用了补充文件4中的药物-靶点关联数据，以及补充文件5中336个宿主蛋白的UniProt登录号（UniProt accession number）。针对Gysi等的相互作用组（Gysi等，2021，DOI: 10.1073/pnas.2025581118），本研究使用了补充文件6中336个宿主蛋白的Entrez ID（Entrez ID）。补充文件7展示了DrugBank数据库（DrugBank）药物的ATC分类（Anatomical Therapeutic Chemical分类）。根据CMAP（Connectivity Map）、体外实验与临床试验证据显示对SARS-CoV-2（严重急性呼吸综合征冠状病毒2）具有作用的DrugBank数据库条目，分别收录于补充文件8、9与10。本研究网络医学方法的预测结果见于补充文件11。基于核方法的宿主蛋白权重分配结果收录于补充文件12。