Selected col and rows not working - RStudio

[Romaindonne]

Hello,

I have run the "plot_dot_by_column_name.r" directly in RStudio after facing many difficulties with Python.
It works, except when I try to add the "selected_columns" and "selected_rows". This is the same when using the tutorial data.

TestTuto = dot_plot (selected_rows = "./out_Tutorial/Test/raw.txt",
selected_columns = "./out_Tutorial/Test/col.txt",
filename = 'Tuto_selected_plot.pdf',
width = 8,
height = 10,
means_path = './out_Tutorial/Test/significant_means.txt',
pvalues_path = './out_Tutorial/Test/pvalues.txt',
means_separator = '\t',
pvalues_separator = '\t',
output_extension = '.pdf' )
I got this error:
Error in data.frame(..., check.names = FALSE) :
arguments imply differing number of rows: 1, 0
Called from: data.frame(..., check.names = FALSE)
Browse[1]> Q

Thanks for your help!

[luzgaral]

Dear Romaindonne,

It looks like your input has a differing number of rows: 1, 0. You might need to amend the input or modify the parameters in the read function accordingly.

Best,

Luz

[Romaindonne]

Dear luzgaral,

Thanks for your help!

I don’t really understand because the two inputs (means and pavalues/relevant_interactions) have been generated with cellphoneDB function in python.It is true that the two files don’t have the same number of cell/cell interactions.

But when I do "dot plot" function in R, there is no problem, it generates the dot plot.

When I try to pick selective rows and columns as mentioned in the tutorial with rows.txt and columns.txt I run the error.

And I run the error with the tutorial exemple.

How can I change the read function?

Thanks

---

dot_plot = function(selected_rows = "raw.txt",
selected_columns = "col.txt",
filename = 'plot.pdf',
width = 8,
height = 10,
means_path = './means.txt',
pvalues_path = './pvalues.txt',
means_separator = '\t',
pvalues_separator = '\t',
output_extension = '.pdf'
){

all_pval = read.table(pvalues_path, header=T, stringsAsFactors = F, sep=means_separator, comment.char = '', check.names=F)
all_means = read.table(means_path, header=T, stringsAsFactors = F, sep=pvalues_separator, comment.char = '', check.names=F)

intr_pairs = all_pval$interacting_pair
all_pval = all_pval[,-c(1:11)]
all_means = all_means[,-c(1:11)]

if(is.null(selected_rows)){
selected_rows = intr_pairs
}

if(is.null(selected_columns)){
selected_columns = colnames(all_pval)
}

sel_pval = all_pval[match(selected_rows, intr_pairs), selected_columns]
sel_means = all_means[match(selected_rows, intr_pairs), selected_columns]

df_names = expand.grid(selected_rows, selected_columns)
pval = unlist(sel_pval)
pval[pval==0] = 0.0009
plot.data = cbind(df_names,pval)
pr = unlist(as.data.frame(sel_means))
pr[pr==0] = 1
plot.data = cbind(plot.data,log2(pr))
colnames(plot.data) = c('pair', 'clusters', 'pvalue', 'mean')

my_palette <- colorRampPalette(c("black", "blue", "yellow", "red"), alpha=TRUE)(n=399)

ggplot(plot.data,aes(x=clusters,y=pair)) +
geom_point(aes(size=-log10(pvalue),color=mean)) +
scale_color_gradientn('Log2 mean (Molecule 1, Molecule 2)', colors=my_palette) +
theme_bw() +
theme(panel.grid.minor = element_blank(),
panel.grid.major = element_blank(),
axis.text=element_text(size=14, colour = "black"),
axis.text.x = element_text(angle = 90, hjust = 1),
axis.text.y = element_text(size=12, colour = "black"),
axis.title=element_blank(),
panel.border = element_rect(size = 0.7, linetype = "solid", colour = "black"))

if (output_extension == '.png') {
ggsave(filename, width = width, height = height, device = png, limitsize=F)
}
else {
ggsave(filename, width = width, height = height, limitsize=F)
}
}