What is the best way to process a metadata file that SEURAT can read?

Question

[biotrekker]

Hello, I have been working with the Seurat pipeline. I want to know what the best way to process a metadata file is. I have been using AddMetaData() from Seurat, but I don't know how to transform my metadata file such that all the cell barcodes are represented, as required by the function. I have attached partial images of alldata and metadata objects.

alldata_m<-AddMetaData(alldata, metadata)

Any help would be appreciated, Thanks

Answer 1

[fracarb8]

The metadata is saved in the meta.data slot at it is just a table, meaning that you can treat it like a normal dataframe.

# create the SampleID column by splitting the rownames 
[e-mail]$SampleID <- gsub("(^.+)_.*$","\\1", rownames([e-mail])
# or [e-mail]$SampleID <- [e-mail]$orig.ident

# add the other columns by matching the values from the other table
exp_meta <- the other dataframe

[e-mail]$patient <- sapply([e-mail]$SampleID, function(ita) exp_meta[match(ita,exp_meta$SampleID),"Patient"])
[e-mail]$Type <- sapply([e-mail]$SampleID, function(ita) exp_meta[match(ita,exp_meta$SampleID),"Type"])

Alternatively you can use merge.data.frame or any other function that join (left/right) the 2 table together.

If you want to use the AddMetaData wrapper

metadata <- [e-mail]
# add all the columns to `metadata`
...
seurat <- AddMetaData(seurat,metadata = metadata)