Violin plot (Monocle 3) - Troubleshooting

Question

[seattlescientist_01]

I am trying to generate some violin plots in monocle on a cell dataset object. Basically, I'm trying to visualize expression of certain marker genes in different clusters of a cell dataset object.

I tried to follow this online documentation using the plot_genes_violin function in monocle3 but it's giving me an error (see attached screenshot). Does anyone have any ideas on how I can troubleshoot this error or if there is a different way to run this in ggplot?

#cell dataset object = cds
cds_subset <- cds[row.names(subset(rowData(cds),
                     gene_short_name %in% c("Podxl", "Nphs1", "Nphs2"))),]
plot_genes_violin(cds_subset)

[Ram]

Please do not paste screenshots of plain text content, it is counterproductive. You should copy paste the content directly here (using the code formatting option shown below).

Answer 1

[bk11]

I have not tested it though. Can you try this-

your_genes=c("Podxl", "Nphs1", "Nphs2")
cds_subset <- cds[rowData(cds)$gene_short_name %in% your_genes,]

plot_genes_violin(cds_subset, group_cells_by="embryo.time.bin", ncol=2) +
      theme(axis.text.x=element_text(angle=45, hjust=1)) #in group_cells_by please choose what you want to group by.

[seattlescientist_01]

I'm still seeing errors:

First I tried the code you mentioned except I use clusters instead of embryo.time.bin since this is a column in my colData(cds).

plot_genes_violin(cds_subset, group_cells_by="clusters", ncol=2) + theme(axis.text.x=element_text(angle=45, hjust=1))
Error in plot_genes_violin(cds_subset, group_cells_by = "clusters", ncol = 2) :  unused argument (group_cells_by = "clusters")

Then, I tried this code again without the group_cells_by portion

plot_genes_violin(cds_subset) 
Error in cds_subset@expressionFamily@vfamily %in% c("negbinomial", "negbinomial.size") :  no slot of name "expressionFamily" for this object of class "cell_data_set"

Is it possible there is a bug in this code?

I'm also wondering if there is a way to generate violin plots using ggplot.

[bk11]

No there is no bug in this. Please see carefully what the error message tells to you.

[seattlescientist_01]

I see the error message but I'm confused on what it's telling me.

[Ram]

What is the output to:

help.search('plot_genes_violin', agrep=FALSE, ignore.case=FALSE, fields=c('name'))
lsf.str()
sessionInfo()

[bk11]

I m not sure why you have problem there. I do not encounter any problem. May be your gene list are not present in your source object. Please see below how I would do it and it works fine. I m using tutorial from monocle3.

library(monocle3)
library(dplyr) # imported for some downstream data manipulation

expression_matrix <- readRDS(url("https://depts.washington.edu:/trapnell-lab/software/monocle3/celegans/data/cao_l2_expression.rds"))
cell_metadata <- readRDS(url("https://depts.washington.edu:/trapnell-lab/software/monocle3/celegans/data/cao_l2_colData.rds"))
gene_annotation <- readRDS(url("https://depts.washington.edu:/trapnell-lab/software/monocle3/celegans/data/cao_l2_rowData.rds"))

cds <- new_cell_data_set(expression_matrix,
                         cell_metadata = cell_metadata,
                         gene_metadata = gene_annotation)
cds
class: cell_data_set 
dim: 20271 42035 
metadata(1): cds_version
assays(1): counts
rownames(20271): WBGene00000001 WBGene00000002 ... WBGene00269394 WBGene00269421
rowData names(1): gene_short_name
colnames(42035): cele-001-001.CATGACTCAA cele-001-001.AAGACGGCCA ...
  cele-010-092.GTATTAAGTT cele-010-092.GAAGTCCGTC
colData names(5): plate cao_cluster cao_cell_type cao_tissue Size_Factor
reducedDimNames(0):
mainExpName: NULL
altExpNames(0):

test_genes <- c("che-1",
                    "hlh-17",
                    "nhr-6",
                    "dmd-6",
                    "ceh-36",
                    "ham-1")
cds_subset <- cds[rowData(cds)$gene_short_name %in% test_genes,]

colData(cds_subset)
    DataFrame with 42035 rows and 5 columns
                           plate cao_cluster          cao_cell_type             cao_tissue Size_Factor
                        <factor>    <factor>            <character>            <character>   <numeric>
cele-001-001.CATGACTCAA      001          20   Unclassified neurons                Neurons    0.236833
cele-001-001.AAGACGGCCA      001          6                Germline                  Gonad    1.299291
cele-001-001.GCCAACGCCA      001          13 Intestinal/rectal mu.. Intestinal/rectal mu..    1.368367
cele-001-001.ATAGGAGTAC      001          27      Vulval precursors                     NA    0.324000
cele-001-001.CTCGTCTAGG      001          2            Coelomocytes           Coelomocytes    0.781219

plot_genes_violin(cds_subset, group_cells_by="plate", ncol=2) +
  theme(axis.text.x=element_text(angle=45, hjust=1)) 

[Ram]

The error message tells me there is the possibility of a conflicting function called plot_genes_violin that accepts one parameter and errors out with the named parameter that the expected plot_genes_violin function accepts without complaint.