Existing blood transcriptional classifiers accurately discriminate active tuberculosis from latent infection in individuals from south India
Existing blood transcriptional classifiers accurately discriminate active tuberculosis from latent infection in individuals from south India
复制标题
DOI：
10.1016/j.tube.2018.01.002
复制
发表时间：
2018-03-01
期刊：
影响因子：
3.2
通讯作者：
Salgame, Padmini
中科院分区：
文献类型：
--
作者：
Leong, Samantha;Zhao, Yue;Salgame, Padmini
几项研究已经确定了可以将活性与潜在结核病（TB）区分开的血液转录组特征。这项研究的目的是评估这些现有基因谱对南印度人群中结核病疾病的分类程度。对从28例结核病患者收集的全血PAXGENE样品和16名受试者（LTBI）受试者进行的全血PAXGENE样品进行了RNA测序，作为正在进行的家庭接触研究的一部分。进行了差异基因表达和聚类分析，并根据已建立的基因特征对TB和LTBI个体的显式预测测试进行了比较。我们基于已知基因集的表达（ROC AUC 0.9007-0.9879）观察到结核病态的强烈预测性能。总之，我们的发现表明，以前报道的从不同种族产生的分类器可以准确地区分南印度患者的LTBI的活性结核病。未来的工作应集中于将现有基因特征转换为通用结核基因签名，以诊断，监测结核病治疗和评估新药物方案。
Several studies have identified blood transcriptomic signatures that can distinguish active from latent Tuberculosis (TB). The purpose of this study was to assess how well these existing gene profiles classify TB disease in a South Indian population. RNA sequencing was performed on whole blood PAXgene samples collected from 28 TB patients and 16 latently TB infected (LTBI) subjects enrolled as part of an ongoing household contact study. Differential gene expression and clustering analyses were performed and compared with explicit predictive testing of TB and LTBI individuals based on established gene signatures. We observed strong predictive performance of TB disease states based on expression of known gene sets (ROC AUC 0.9007-0.9879). Together, our findings indicate that previously reported classifiers generated from different ethnic populations can accurately discriminate active TB from LTBI in South Indian patients. Future work should focus on converting existing gene signatures into a universal TB gene signature for diagnosis, monitoring TB treatment, and evaluating new drug regimens.