从DESEQ2设计中删除大脑区域变化
相当陌生，所以我很感激您的耐心！
因此，我使用RNNA-seq基因表达式计算了从GTEX下载的数据，它使用Re：Recout3 package in R：
gtex_rse <- recount3::create_rse_manual(
project = "BRAIN",
project_home = "data_sources/gtex",
organism = "human",
annotation = "gencode_v26",
type = "gene"
)
与TCGA患者肿瘤组织数据进行比较，但TCGA数据并未指定收集的大脑区域。因此，我想通过肿瘤与控制表达差异进行比较，而不会影响大脑区域。是否有任何建议最小化大脑区域对肿瘤与控制差异表达分析的影响？提前致谢！！
Fairly new to this so I would appreciate your patience!
So I have RNA-seq gene expression counts data downloaded from GTEx using the recount3 package in r:
gtex_rse <- recount3::create_rse_manual(
project = "BRAIN",
project_home = "data_sources/gtex",
organism = "human",
annotation = "gencode_v26",
type = "gene"
)
And there is definitely some variation attributable to brain region (this is including just sex in the design):
I want to compare this expression data to TCGA patient tumor tissue data, but the TCGA data does not specify the brain region of collection. Thus, I'd like to compare by tumor vs control expression differences without influence of brain region. Are there any suggestions for minimizing influence of brain region on tumor vs control differential expression analysis? Thanks in advance!!
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
原则上，按颜色编码对组的可视化不应影响聚类本身。但是，跨大脑区域的基因表达谱可能存在差异，而大脑区域固有的细胞群体可能存在差异。这就是为什么在数据库中进行比较可能会很棘手的原因，因为这些样品可以通过实验方式进行不同的方式制备。虽然，您可以尝试删除大脑标签列，并查看是否存在差异。
In principle, the visualization of group by color coding should not affect the clustering per se. However, there could be differences in gene expression profiles across brain regions that are inherent their to cell populations. This is why comparing across databases can be tricky since the samples could have been prepared in different ways experimentally. Although, you can try deleting the brain label column and see whether there is a difference.