PDF(Pubmed)
Abstract：
UNASSIGNED: Increasing evidence reveals the involvement of mitochondria and macrophage polarisation in tumourigenesis and progression. This study aimed to establish mitochondria and macrophage polarisation-associated molecular signatures to predict prognosis in gastric cancer (GC) by single-cell and transcriptional data.
UNASSIGNED: Initially, candidate genes associated with mitochondria and macrophage polarisation were identified by differential expression analysis and weighted gene co-expression network analysis. Subsequently, candidate genes were incorporated in univariateCox analysis and LASSO to acquire prognostic genes in GC, and risk model was created. Furthermore, independent prognostic indicators were screened by combining risk score with clinical characteristics, and a nomogram was created to forecast survival in GC patients. Further, in single-cell data analysis, cell clusters and cell subpopulations were yielded, followed by the completion of pseudo-time analysis. Furthermore, a more comprehensive immunological analysis was executed to uncover the relationship between GC and immunological characteristics. Ultimately, expression level of prognostic genes was validated through public datasets and qRT-PCR.
UNASSIGNED: A risk model including six prognostic genes (GPX3, GJA1, VCAN, RGS2, LOX, and CTHRC1) associated with mitochondria and macrophage polarisation was developed, which was efficient in forecasting the survival of GC patients. The GC patients were categorized into high-/low-risk subgroups in accordance with median risk score, with the high-risk subgroup having lower survival rates. Afterwards, a nomogram incorporating risk score and age was generated, and it had significant predictive value for predicting GC survival with higher predictive accuracy than risk model. Immunological analyses revealed showed higher levels of M2 macrophage infiltration in high-risk subgroup and the strongest positive correlation between risk score and M2 macrophages. Besides, further analyses demonstrated a better outcome for immunotherapy in low-risk patients. In single-cell and pseudo-time analyses, stromal cells were identified as key cells, and a relatively complete developmental trajectory existed for stromal C1 in three subclasses. Ultimately, expression analysis revealed that the expression trend of RGS2, GJA1, GPX3, and VCAN was consistent with the results of the TCGA-GC dataset.
UNASSIGNED: Our findings demonstrated that a novel prognostic model constructed in accordance with six prognostic genes might facilitate the improvement of personalised prognosis and treatment of GC patients.
UNASSIGNED: Initially, candidate genes associated with mitochondria and macrophage polarisation were identified by differential expression analysis and weighted gene co-expression network analysis. Subsequently, candidate genes were incorporated in univariateCox analysis and LASSO to acquire prognostic genes in GC, and risk model was created. Furthermore, independent prognostic indicators were screened by combining risk score with clinical characteristics, and a nomogram was created to forecast survival in GC patients. Further, in single-cell data analysis, cell clusters and cell subpopulations were yielded, followed by the completion of pseudo-time analysis. Furthermore, a more comprehensive immunological analysis was executed to uncover the relationship between GC and immunological characteristics. Ultimately, expression level of prognostic genes was validated through public datasets and qRT-PCR.
UNASSIGNED: A risk model including six prognostic genes (GPX3, GJA1, VCAN, RGS2, LOX, and CTHRC1) associated with mitochondria and macrophage polarisation was developed, which was efficient in forecasting the survival of GC patients. The GC patients were categorized into high-/low-risk subgroups in accordance with median risk score, with the high-risk subgroup having lower survival rates. Afterwards, a nomogram incorporating risk score and age was generated, and it had significant predictive value for predicting GC survival with higher predictive accuracy than risk model. Immunological analyses revealed showed higher levels of M2 macrophage infiltration in high-risk subgroup and the strongest positive correlation between risk score and M2 macrophages. Besides, further analyses demonstrated a better outcome for immunotherapy in low-risk patients. In single-cell and pseudo-time analyses, stromal cells were identified as key cells, and a relatively complete developmental trajectory existed for stromal C1 in three subclasses. Ultimately, expression analysis revealed that the expression trend of RGS2, GJA1, GPX3, and VCAN was consistent with the results of the TCGA-GC dataset.
UNASSIGNED: Our findings demonstrated that a novel prognostic model constructed in accordance with six prognostic genes might facilitate the improvement of personalised prognosis and treatment of GC patients.
摘要：
■越来越多的证据表明线粒体和巨噬细胞极化参与肿瘤的发生和发展。这项研究旨在建立线粒体和巨噬细胞极化相关的分子特征，以通过单细胞和转录数据预测胃癌（GC）的预后。
■最初，通过差异表达分析和加权基因共表达网络分析鉴定与线粒体和巨噬细胞极化相关的候选基因.随后，将候选基因纳入单变量Cox分析和LASSO以获得GC中的预后基因，并创建了风险模型。此外，通过结合风险评分和临床特征筛选独立的预后指标,并创建列线图来预测GC患者的生存率。Further,在单细胞数据分析中，产生了细胞簇和细胞亚群，然后完成伪时间分析。此外，进行了更全面的免疫学分析，以揭示GC与免疫学特征之间的关系。最终,通过公开数据集和qRT-PCR验证预后基因的表达水平。
■包括六个预后基因（GPX3，GJA1，VCAN，RGS2,LOX,和CTHRC1）与线粒体和巨噬细胞极化相关，这在预测GC患者的生存率方面是有效的。根据中位风险评分将GC患者分为高/低风险亚组，高风险亚组的生存率较低。之后,生成了包含风险评分和年龄的列线图，与风险模型相比，它对预测GC生存具有显著的预测价值。免疫学分析显示，高风险亚组中M2巨噬细胞浸润水平较高，风险评分与M2巨噬细胞之间的正相关最强。此外,进一步分析显示,在低危患者中,免疫治疗的结局更好.在单细胞和伪时间分析中，基质细胞被确定为关键细胞，基质C1在三个亚类中存在相对完整的发育轨迹。最终,表达分析显示，RGS2，GJA1，GPX3和VCAN的表达趋势与TCGA-GC数据集的结果一致。
■我们的发现表明，根据六个预后基因构建的新型预后模型可能有助于改善GC患者的个性化预后和治疗。
■最初，通过差异表达分析和加权基因共表达网络分析鉴定与线粒体和巨噬细胞极化相关的候选基因.随后，将候选基因纳入单变量Cox分析和LASSO以获得GC中的预后基因，并创建了风险模型。此外，通过结合风险评分和临床特征筛选独立的预后指标,并创建列线图来预测GC患者的生存率。Further,在单细胞数据分析中，产生了细胞簇和细胞亚群，然后完成伪时间分析。此外，进行了更全面的免疫学分析，以揭示GC与免疫学特征之间的关系。最终,通过公开数据集和qRT-PCR验证预后基因的表达水平。
■包括六个预后基因（GPX3，GJA1，VCAN，RGS2,LOX,和CTHRC1）与线粒体和巨噬细胞极化相关，这在预测GC患者的生存率方面是有效的。根据中位风险评分将GC患者分为高/低风险亚组，高风险亚组的生存率较低。之后,生成了包含风险评分和年龄的列线图，与风险模型相比，它对预测GC生存具有显著的预测价值。免疫学分析显示，高风险亚组中M2巨噬细胞浸润水平较高，风险评分与M2巨噬细胞之间的正相关最强。此外,进一步分析显示,在低危患者中,免疫治疗的结局更好.在单细胞和伪时间分析中，基质细胞被确定为关键细胞，基质C1在三个亚类中存在相对完整的发育轨迹。最终,表达分析显示，RGS2，GJA1，GPX3和VCAN的表达趋势与TCGA-GC数据集的结果一致。
■我们的发现表明，根据六个预后基因构建的新型预后模型可能有助于改善GC患者的个性化预后和治疗。
