INTRODUCTION Although not considered a primary cause, neuroinflammation is associated with many neurodegenerative disorders, including Parkinson's disease (PD). METHODS To elucidate potential immune involvement in PD, the present study imputed immune cell abundances from bulk RNA-sequencing transcriptomic data of PD postmortem prefrontal cortices. CIBERSORTx, an RNA deconvolution algorithm that implements support vector regression, was used to measure the relative abundances of immune cells from a previously published gene expression dataset. Through this machine-learning approach, relative proportions of 22 immune cell subtypes present in the original brain tissue were estimated. RESULTS Prefrontal cortices from PD patients exhibited significantly higher relative abundances of monocytes compared to neuropathologically normal controls (p value = 0.0005). The relative proportions of the other 21 immune subtypes showed no significant differences between control and PD samples. CONCLUSION AND DISCUSSION The findings corroborate previous reports and suggest monocytes may be involved in PD pathogenesis.

中文翻译：

---

前额叶皮层转录组解卷积涉及帕金森病中的单核细胞浸润

引言 尽管不被认为是主要原因，但神经炎症与许多神经退行性疾病有关，包括帕金森病 (PD)。方法 为了阐明 PD 中潜在的免疫参与，本研究从 PD 死后前额叶皮质的大量 RNA 测序转录组数据中估算了免疫细胞丰度。CIBERSORTx 是一种实现支持向量回归的 RNA 解卷积算法，用于测量来自先前发布的基因表达数据集的免疫细胞的相对丰度。通过这种机器学习方法，估计了原始脑组织中存在的 22 种免疫细胞亚型的相对比例。结果 与神经病理学正常对照相比，PD 患者的前额叶皮质表现出显着更高的单核细胞相对丰度（p 值 = 0.0005）。其他 21 种免疫亚型的相对比例在对照和 PD 样品之间没有显着差异。结论与讨论 这些发现证实了之前的报道，并表明单核细胞可能参与了 PD 的发病机制。