ixxmu / mp_duty

抓取网络文章到github issues保存
https://archives.duty-machine.now.sh/
105 stars 30 forks source link

人类衰老状态的血液免疫单细胞图谱 #2188

Closed ixxmu closed 2 years ago

ixxmu commented 2 years ago

https://mp.weixin.qq.com/s/rUx5CfCMC_x-oXFooD0lbw

github-actions[bot] commented 2 years ago

人类衰老状态的血液免疫单细胞图谱 by 生信技能树

文章:《Multidimensional single-cell analysis of human peripheral blood reveals characteristic features of the immune system landscape in aging and frailty》

数据集是 GSE157007 ,共产生了 high-quality scRNA-seq data from 114,467 mononuclear cells

样品队列

样品分成4个分组

可以看到, 样品分成4个分组,cord blood, young adults and healthy and frail old adults  :

  • 新生儿脐带血(n=3)
  • 青年组PBMCs(n=3)
  • 健康老年组PBMCs(n=6)
  • 体弱老年组PBMCs(n=5)

3个单细胞技术 :

  • single-cell RNA sequencing (scRNA-seq),
  • T-cell antigen receptor (TCR) repertoire sequencing (scTCR-seq)
  • surface protein antibody-barcode sequencing (CCR7, CD45RA, CD4+ and CD8+)

第一层次降维聚类分群

初步分成了 17 clusters ,大家在进行血液单细胞免疫亚群细分的时候可以参考 :

  • Clusters 1, 2 and 3 were distinguishable as naïve, memory CD4+ and memory CD8+ T cells,
  • Cluster 6 and 9 with mixed expression of CD4+ and CD8+ were named as ‘other T cells’
  • Clusters 7 and 16 (CD79A, CD74 and MS4A1 high) were identified as B cells and antigen presenting B (APC B) cells, respectively
  • Clusters 5, 8 and 14 (NKG7, GNLY, KLRB1 and GZMB high) were identified as natural killer (NK) cells.
  • The myeloid lineage cells (clusters 4, 10 and 12) were classified as classical, intermediate and nonclassical monocytes, respectively, attributing to varying levels of S100A9, LYZ, IL1B and FCGR3A expression.
  • myeloid dendritic cells (mDCs; cluster 11),
  • plasmacytoid DCs (pDCs; cluster 15),
  • platelets (cluster 13)
  • granulocytes (cluster 17)

基本上跟我们一直给大家的单细胞亚群标记基因差不多了。

初步分成了 17 clusters

接下来就是对每个单细胞亚群进行细分了,比如T细胞,可以看到跟肿瘤微环境单细胞转录组的T细胞构成很不一样,没有大量的耗竭相关t细胞 :

T细胞细分亚群

因为有4个分组,cord blood, young adults and healthy and frail old adults  ,所以可以看不同组的不同单细胞亚群的比例情况差异,如下所示 。

细胞亚群的比例情况差异

可以看到规律是:

  • B细胞占比在各年龄组中相对稳定
  • 单核细胞在青年组中占比较低,在新生组和老年组占比均较高
  • 新生组样本中,包含大量的Naïve T细胞和极少量的Memory T细胞
  • Naïve T细胞的比例随着年龄的增长而呈现出减少的趋势,从青年组过渡至老年组时,Naïve T细胞的数量更是急剧下降
  • 体弱老年组CD4+T CM细胞较多,健康老年组CD8+T CM和Treg细胞较多。这可能意味着CD8+和CD4+T CM细胞之间的平衡在维持老年人免疫系统的活力中起着重要作用

因为还有scTCR-seq,所以需要跟这个普通单细胞转录组表达量矩阵数据分析结合起来,也是同样的看不同分组样品的差异即可。

大家可以去看看OSCA单细胞数据分析


单细胞的多组对照设计(例如正常组与给药组)可以为细胞类型水平比较提供以往Bulk RNA-seq分析所不能达到的精度。对此一般有两种进阶分析思路:

  • (1)DE(Differential expression)--两组样本的同一细胞类型的基因表达差异分析;
  • (2)DA(Differential abundance)--两组样本的同一细胞类型的丰度差异分析

写在文末

我在《生信技能树》,《生信菜鸟团》,《单细胞天地》的大量推文教程里面共享的代码都是复制粘贴即可使用的, 有任何疑问欢迎留言讨论,也可以发邮件给我,详细描述你遇到的困难的前因后果给我,我的邮箱地址是 jmzeng1314@163.com

如果你确实觉得我的教程对你的科研课题有帮助,让你茅塞顿开,或者说你的课题大量使用我的技能,烦请日后在发表自己的成果的时候,加上一个简短的致谢,如下所示:

We thank Dr.Jianming Zeng(University of Macau), and all the members of his bioinformatics team, biotrainee, for generously sharing their experience and codes.

十年后我环游世界各地的高校以及科研院所(当然包括中国大陆)的时候,如果有这样的情谊,我会优先见你。