I have downloaded a scRNA-seq matirx. As the author had annotated or clusted the cells. so I want to read this matirx with seurat and check the gene expression in feature plot. Does anyone know how to do it? Thanks in advance! Best,
Hi,
If you want to do a feature plot you should have a UMAP or TSNE of clusters. So you need to make a Seurat object from the matrix, remove cells with low features, normalize the data, scale it, find variable features, run PCA analysis, find neighbors, then find clusters, finally fillowed by running a UMAP or TSNE to visualize the clusters. Then using the UMAP or TSNE object do the feature plot.
FeaturePlot(<your object>, features = "<your features>")
Follow https://satijalab.org/seurat/v3.1/pbmc3k_tutorial.html
and
https://satijalab.org/seurat/v3.1/visualization_vignette.html
from the authors of Seurat.
Best, Dan
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
What have you tried? The Seurat docs give pretty explicit directions.
what I want to do is to just make the feature plots foe some genes, not to reanalyzed the scRNA-seq data.
Okay. So what have you tried? Creating a Seurat object (from which feature plots can be created) is not particularly difficult. We can't help you learn if you don't show us what the sticking points are.
no. before making a feature plot, I should cluster the cells. if I do not clsuter the cell, I should not make the feature plot, right? As I just want to make a feature plot.
Although I creat a seurat object, I should not skip several steps, such as cluster the cells, right?
You said you already have clusters. Just set them as the identities of each cell - see
?Ident. Depends on the matrix as to whether you should normalize it.