SplitObject() function not recognizing my column to split by?

[sarakn97]

I am trying to split my seurat object based off of my "sample" column which contains the values "Infected" or "Not". I have run this exact code several times in the past and it works yet today it does not. Below is the code and the error:

split_seurat <- split_seurat[c("Infected", "Not")]

Error:

Error in intI(i, n = d[1], dn[[1]], give.dn = FALSE) : 
  invalid character indexing

As you can see, the sample column does exist and I am able to select it in other functions

                       orig.ident nCount_RNA nFeature_RNA   sample percent.mt
Infected_AAACCCACAAGACGGT-1 BM_infected      10937         3996 Infected  1.9292311
Not_TTTGGTTTCGTAGGGA-1  NotInf_BM      39308         7177    Not   4.273939

Previous Code:

BM_Infected <- read.csv("/storage1/fs1//Active/scRNA_sara/scRNA_BM_sara/BM_I/gem_classification.csv")
BM_Infected_mouse <- subset(BM_Infected, BM_Infected$call=="mm10")

BM_NOTinfected <- read.csv("/storage1/fs1/Active/scRNA_sara/scRNA_BM_sara/BM_NI/gem_classification.csv")
BM_NOTinfected_mouse <- subset(BM_NOTinfected, BM_NOTinfected$call=="mm10")

# read in sample
BM_HIV = Read10X("/storage1/fs1/Active/scRNA_sara/scRNA_BM_sara/BM_I/filtered_feature_bc_matrix/")
dim(BM_HIV)
#68886  #5356

# remove mouse cells (BM_HIV)  
BM_HIV <- BM_HIV[,!colnames(BM_HIV)%in%BM_Infected_mouse$barcode]
dim(BM_HIV)
#68886   #5318

# Create Sample1 Seurat Object (Infected)
infected_BM <- CreateSeuratObject(counts = BM_HIV, project = "BM_infected", min.cells = 3, min.features = 200)
dim(infected_BM)
#24286 5286
head(infected_BM)

# read sample 2
BM_NI = Read10X("/storage1/fs1/Active/scRNA_sara/scRNA_BM_sara/BM_NI/filtered_feature_bc_matrix/")
dim(BM_NI)
#68886  7553

# remove mouse cells (BM_NI) 
BM_NI <- BM_NI[,!colnames(BM_NI)%in%BM_NOTinfected_mouse$barcode]
dim(BM_NI)
#68886 7529

# create sample2 suerat object
NotInf_BM <- CreateSeuratObject(counts = BM_NI, project = "NotInf_BM", min.cells = 3, min.features = 200)
dim(NotInf_BM)
# 29325  7428
head(NotInf_BM)

######################################################################################
# Identify Condition
infected_BM[["sample"]] <- "Infected"
NotInf_BM[["sample"]] <- "Not"

# Merge Samples
merged_seurat <- merge(x=infected_BM, y=NotInf_BM, add.cell.id = c("Infected", "Not"))

sum(grepl("MT",rownames(merged_seurat)))
# 254
rownames(merged_seurat)[grepl("MT-",rownames(merged_seurat))]

# calculate the proportion of transcripts mapping to mitochondrial genes
merged_seurat[["percent.mt"]] <- PercentageFeatureSet(merged_seurat, pattern = "MT-")

# Create metadata dataframe
metadata <- [email protected]
# Add cell IDs to metadata
metadata$cells <- rownames(metadata)

#save(merged_seurat, file="./merged_BM_seurat.RData")
#load("./merged_BM_seurat.RData")
max([email protected]$percent.mt)
#84.3038

# PLot Quality
VlnPlot(merged_seurat, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"), ncol = 3)

# Filter based on quality plots 
Filter_BM <- subset(x=merged_seurat, subset=nFeature_RNA > 200 & nFeature_RNA < 9000 & percent.mt > -Inf & percent.mt <6)
VlnPlot(Filter_BM, features=c("nFeature_RNA", "nCount_RNA", "percent.mt"), ncol=3)


# Sums post Filtering
sum(Filter_BM$sample=="Infected")
# 5062
sum(Filter_BM$sample=="Not")
# 6842

Filter_BM@assays$RNA@counts@Dimnames[[1]] <- gsub("GRCh38filtered_feature_bc_matrix-","",Filter_BM@assays$RNA@counts@Dimnames[[1]])
Filter_BM@assays$RNA@data@Dimnames[[1]] <- gsub("GRCh38-","",Filter_BM@assays$RNA@counts@Dimnames[[1]])

Not sure what is going on and why it now does not work.

[samuel-marsh]

Hi,

You don't appear to actually be using the SplitObject function in the code you posted. What is happens if you run:

obj_list <- SplitObject(object = COMB_OBJ, split.by = "sample")

Best,
Sam