Department of Mechatronics Engineering, Jeju National University, Jeju, South Korea.
BioSpero, Inc., Jeju, South Korea.
Front Public Health. 2022 Jun 16;10:902123. doi: 10.3389/fpubh.2022.902123. eCollection 2022.
The global spread of the SARS coronavirus 2 (SARS-CoV-2), its manifestation in human hosts as a contagious disease, and its variants have induced a pandemic resulting in the deaths of over 6,000,000 people. Extensive efforts have been devoted to drug research to cure and refrain the spread of COVID-19, but only one drug has received FDA approval yet. Traditional drug discovery is inefficient, costly, and unable to react to pandemic threats. Drug repurposing represents an effective strategy for drug discovery and reduces the time and cost compared to drug discovery. In this study, a generic drug repurposing framework (SperoPredictor) has been developed which systematically integrates the various types of drugs and disease data and takes the advantage of machine learning (Random Forest, Tree Ensemble, and Gradient Boosted Trees) to repurpose potential drug candidates against any disease of interest. Drug and disease data for FDA-approved drugs ( = 2,865), containing four drug features and three disease features, were collected from chemical and biological databases and integrated with the form of drug-disease association tables. The resulting dataset was split into 70% for training, 15% for testing, and the remaining 15% for validation. The testing and validation accuracies of the models were 99.3% for Random Forest and 99.03% for Tree Ensemble. In practice, SperoPredictor identified 25 potential drug candidates against 6 human host-target proteomes identified from a systematic review of journals. Literature-based validation indicated 12 of 25 predicted drugs (48%) have been already used for COVID-19 followed by molecular docking and re-docking which indicated 4 of 13 drugs (30%) as potential candidates against COVID-19 to be pre-clinically and clinically validated. Finally, SperoPredictor results illustrated the ability of the platform to be rapidly deployed to repurpose the drugs as a rapid response to emergent situations (like COVID-19 and other pandemics).
SARS 冠状病毒 2（SARS-CoV-2）在全球范围内的传播、在人类宿主中的表现为传染病以及其变体引发了大流行，导致超过 600 万人死亡。人们投入了大量精力进行药物研究以治疗和阻止 COVID-19 的传播，但目前仅有一种药物获得了 FDA 的批准。传统的药物发现方法效率低下、成本高昂，且无法应对大流行的威胁。药物再利用代表了一种有效的药物发现策略，与药物发现相比，它可以减少时间和成本。在这项研究中，开发了一种通用的药物再利用框架（SperoPredictor），该框架系统地整合了各种类型的药物和疾病数据，并利用机器学习（随机森林、树集成和梯度提升树）来重新利用针对任何感兴趣疾病的潜在药物候选物。从化学和生物数据库中收集了 FDA 批准药物（=2865 种）的药物和疾病数据，这些药物包含四种药物特征和三种疾病特征，并将其整合为药物-疾病关联表的形式。将得到的数据集分为 70%用于训练、15%用于测试，其余 15%用于验证。随机森林模型的测试和验证准确率为 99.3%，树集成模型的测试和验证准确率为 99.03%。在实际应用中，SperoPredictor 从系统综述期刊中鉴定了 6 个人类宿主靶标蛋白质组，识别出 25 种针对这些蛋白质组的潜在药物候选物。基于文献的验证表明，25 种预测药物中有 12 种（48%）已用于 COVID-19，随后进行分子对接和重新对接，结果表明其中 13 种药物中的 4 种（30%）可能是针对 COVID-19 的候选药物，有待进行临床前和临床验证。最后，SperoPredictor 的结果表明，该平台能够迅速部署，重新利用药物，以快速应对紧急情况（如 COVID-19 和其他大流行）。
