Analysis of gene expression in rheumatoid arthritis and related conditions offers insights into sex-bias, gene biotypes and co-expression patterns

The era of next-generation sequencing has mounted the foundation of many gene expression studies. In rheumatoid arthritis research, this has led to the discovery of important candidate genes which offered novel insights into mechanisms and their possible roles in the cure of the disease. In the last years, data generation has outstripped data analysis and while many studies focused on specific aspects of the disease, a global picture of the disease is not yet accomplished. Here, we analyzed and compared a collection of gene expression information from healthy individuals and from patients suffering under different arthritis conditions from published studies containing the following clinical conditions: early and established rheumatoid arthritis, osteoarthritis and arthralgia. We show comprehensive overviews of this data collection and give new insights specifically on gene expression in the early stage, into sex-dependent gene expression, and we describe general differences in expression of different biotypes of genes. Many genes that are related to cytoskeleton changes (actin filament related genes) are differently expressed in early rheumatoid arthritis in comparison to healthy subjects; interestingly, eight of these genes reverse their expression ratio significantly between men and women compared early rheumatoid arthritis and healthy subjects. There are some slighter changes between men and woman between the conditions early and established rheumatoid arthritis. Another aspect are miRNAs and other gene biotypes which are not only promising candidates for diagnoses but also change their expression grossly in average at rheumatoid arthritis and arthralgia compared to the healthy condition. With a selection of intersecting genes, we were able to generate simple classification models to distinguish between healthy and rheumatoid arthritis as well as between early rheumatoid arthritis to other arthritides based on gene expression.下一代测序（next-generation sequencing）技术时代为众多基因表达研究奠定了坚实基础。在类风湿关节炎（rheumatoid arthritis）研究领域，这一技术推动了关键候选基因的发现，为阐明疾病发病机制及其在疾病治疗中的潜在作用提供了全新见解。近年来，数据产出速率已远超数据分析能力；尽管诸多研究聚焦于该疾病的特定层面，但目前仍未形成对该疾病的全局认知。本研究对已发表文献中的基因表达数据集进行了整合分析与比较，数据集涵盖健康个体以及罹患不同关节炎病症的患者样本，所纳入的临床病症包括早期类风湿关节炎、确诊类风湿关节炎、骨关节炎与关节痛。本研究对该数据集进行了全面梳理，并取得多项新发现：包括早期阶段的基因表达特征、性别依赖型基因表达模式，以及不同基因生物型（gene biotypes）的表达差异概况。与健康个体相比，早期类风湿关节炎患者体内诸多与细胞骨架重塑相关的基因（肌动蛋白丝相关基因（actin filament related genes））呈现差异表达；值得注意的是，在对比早期类风湿关节炎与健康个体的样本时，其中8个基因的表达比值在男性与女性群体中发生了显著反转。在早期与确诊类风湿关节炎患者群体间，男性与女性的基因表达仅存在轻微差异。另一研究重点为微小RNA（miRNAs）及其他基因生物型：它们不仅是颇具前景的诊断候选标志物，且相较于健康状态，类风湿关节炎与关节痛患者体内此类分子的平均表达水平均发生显著改变。通过筛选交集基因，本研究构建了简易分类模型，可基于基因表达特征区分健康个体与类风湿关节炎患者，同时也能区分早期类风湿关节炎与其他关节炎病症。