DataSheet1_Diagnostic signature, subtype classification, and immune infiltration of key m6A regulators in osteomyelitis patients.ZIP

Background: As a recurrent inflammatory bone disease, the treatment of osteomyelitis is always a tricky problem in orthopaedics. N6-methyladenosine (m6A) regulators play significant roles in immune and inflammatory responses. Nevertheless, the function of m6A modification in osteomyelitis remains unclear.
Methods: Based on the key m6A regulators selected by the GSE16129 dataset, a nomogram model was established to predict the incidence of osteomyelitis by using the random forest (RF) method. Through unsupervised clustering, osteomyelitis patients were divided into two m6A subtypes, and the immune infiltration of these subtypes was further evaluated. Validating the accuracy of the diagnostic model for osteomyelitis and the consistency of clustering based on the GSE30119 dataset.
Results: 3 writers of Methyltransferase-like 3 (METTL3), RNA-binding motif protein 15B (RBM15B) and Casitas B-lineage proto-oncogene like 1 (CBLL1) and three readers of YT521-B homology domain-containing protein 1 (YTHDC1), YT521-B homology domain-containing family 3 (YTHDF2) and Leucine-rich PPR motif-containing protein (LRPPRC) were identified by difference analysis, and their Mean Decrease Gini (MDG) scores were all greater than 10. Based on these 6 significant m6A regulators, a nomogram model was developed to predict the incidence of osteomyelitis, and the fitting curve indicated a high degree of fit in both the test and validation groups. Two m6A subtypes (cluster A and cluster B) were identified by the unsupervised clustering method, and there were significant differences in m6A scores and the abundance of immune infiltration between the two m6A subtypes. Among them, two m6A regulators (METTL3 and LRPPRC) were closely related to immune infiltration in patients with osteomyelitis.
Conclusion: m6A regulators play key roles in the molecular subtypes and immune response of osteomyelitis, which may provide assistance for personalized immunotherapy in patients with osteomyelitis.

背景：骨髓炎作为一种复发性炎症性骨病，其治疗始终是骨科领域的棘手难题。N6-甲基腺嘌呤（N6-methyladenosine，m6A）调控因子在免疫与炎症应答中发挥重要作用，然而m6A修饰在骨髓炎中的功能仍未明确。
方法：基于GSE16129数据集筛选得到的关键m6A调控因子，本研究采用随机森林（random forest，RF）法构建列线图模型以预测骨髓炎的发病风险。通过无监督聚类将骨髓炎患者划分为两种m6A亚型，并进一步评估各亚型的免疫浸润情况。此外，基于GSE30119数据集验证骨髓炎诊断模型的准确性以及聚类结果的一致性。
结果：通过差异分析筛选得到3种m6A写入酶（甲基转移酶样3（Methyltransferase-like 3，METTL3）、RNA结合基序蛋白15B（RNA-binding motif protein 15B，RBM15B）以及Casitas B谱系原癌基因样1（Casitas B-lineage proto-oncogene like 1，CBLL1））和3种m6A读取蛋白（YT521-B同源域包含蛋白1（YT521-B homology domain-containing protein 1，YTHDC1）、YT521-B同源域家族3（YT521-B homology domain-containing family 3，YTHDF2）以及富含亮氨酸的PPR基序蛋白（Leucine-rich PPR motif-containing protein，LRPPRC）），其平均基尼指数减少量（Mean Decrease Gini，MDG）得分均大于10。基于这6个具有显著意义的m6A调控因子，本研究构建了骨髓炎发病风险预测的列线图模型，拟合曲线显示该模型在测试组与验证组中均具有较高的拟合度。通过无监督聚类方法鉴定出两种m6A亚型（簇A与簇B），两种亚型的m6A评分及免疫浸润丰度均存在显著差异，其中METTL3与LRPPRC这两种m6A调控因子与骨髓炎患者的免疫浸润密切相关。
结论：m6A调控因子在骨髓炎的分子亚型形成与免疫应答中发挥关键作用，可为骨髓炎患者的个性化免疫治疗提供辅助依据。