Demo datasets for the protocol to identify shared transcriptional risks between diseases and compounds predicted to result in mutual benefit

We present a computational protocol (https://github.com/ghbore/protocol-cancer-cvd-similarity), implemented as a Snakemake workflow, that was used in previous works (Gao et al., 2022; Baylis et al., 2023). This protocol allows researchers to identify shared transcriptional processes that drive disease and to screen existing compounds for mutual benefit. The protocol also includes a description of the pharmacovigilance study design used to validate the effect of novel compounds using electronic health records, where applicable. This repository bundles the datasets used in previous works as an example to run through the Snakemake workflow. These datasets include the TCGA cancer dataset, the STARNET and BiKE CVD datasets, and other dependent resources.本研究提出一种以Snakemake工作流实现的计算流程，其开源仓库地址为https://github.com/ghbore/protocol-cancer-cvd-similarity，此前已在Gao等人2022年、Baylis等人2023年的研究中得到应用。该流程可帮助研究人员识别驱动疾病的共通转录过程，并筛选可使双方获益的现有化合物。本流程还包含了药物警戒（Pharmacovigilance）研究设计的相关说明，该设计用于在适用场景下基于电子健康记录（Electronic Health Records）验证新型化合物的效应。本仓库收录了此前各项研究中使用的数据集，作为可运行该Snakemake工作流的示例数据。这些数据集包括癌症基因组图谱（The Cancer Genome Atlas, TCGA）癌症数据集、STARNET与BiKE心血管疾病（Cardiovascular Disease, CVD）数据集，以及其他相关依赖资源。