How to customize the order of groups within the FeatureStatPlot function

[jphe]

Hi,

I am currently using the FeatureStatPlot function from the SCP package. I would like to know how to customize the order of groups within this function. I have tried using the method to change the order in Seurat, but it doesn't seem to work.

bellow is my code:

[email protected] <- [email protected]$celltype 
levels(adata)

tmp <- adata
My_levels <- c('2C','Morula','ICM','TE','Epi','PrE','VE')
Idents(tmp) <- factor(Idents(tmp), levels= My_levels)

levels(Idents(tmp))

[hermespara]

You need to change the order of factors directly in the metadata column.

[email protected]$celltype <- factor([email protected]$celltype, levels = c('2C','Morula','ICM','TE','Epi','PrE','VE'))
FeatureStatPlot(pancreas_sub, stat.by = c("Your Variable to plot"), group.by = "celltype")

[jphe]

Thanks a lot for your reply, I have tried the code above it works well.

However, it seems not compatible with the bg.by option, it reports an error as bellow.

if omit the bg.by='celltyp' option, the function works without any issues.

FeatureStatPlot(
  srt = tmp, group.by = "celltype", bg.by='celltype',
  stat.by =c("Zscan4c"), add_box = TRUE,
)

Error in colnames(df_table)[apply(df_table, 1, function(x) which(x > 0))]: 类别为'list'的下标不对
Traceback:

1. FeatureStatPlot(srt = tmp, group.by = "celltype", bg.by = "celltype", 
 .     stat.by = c("MERVL-int"), add_box = TRUE, )
2. ExpressionStatPlot(exp.data = exp.data, meta.data = meta.data, 
 .     stat.by = stat.by, group.by = group.by, split.by = split.by, 
 .     bg.by = bg.by, plot.by = "group", fill.by = fill.by, cells = cells, 
 .     keep_empty = keep_empty, individual = individual, plot_type = plot_type, 
 .     palette = palette, palcolor = palcolor, alpha = alpha, bg_palette = bg_palette, 
 .     bg_palcolor = bg_palcolor, bg_alpha = bg_alpha, add_box = add_box, 
 .     box_color = box_color, box_width = box_width, box_ptsize = box_ptsize, 
 .     add_point = add_point, pt.color = pt.color, pt.size = pt.size, 
 .     pt.alpha = pt.alpha, jitter.width = jitter.width, jitter.height = jitter.height, 
 .     add_trend = add_trend, trend_color = trend_color, trend_linewidth = trend_linewidth, 
 .     trend_ptsize = trend_ptsize, add_stat = add_stat, stat_color = stat_color, 
 .     stat_size = stat_size, stat_stroke = stat_stroke, stat_shape = stat_shape, 
 .     add_line = add_line, line_color = line_color, line_size = line_size, 
 .     line_type = line_type, cells.highlight = cells.highlight, 
 .     cols.highlight = cols.highlight, sizes.highlight = sizes.highlight, 
 .     alpha.highlight = alpha.highlight, calculate_coexp = calculate_coexp, 
 .     same.y.lims = same.y.lims, y.min = y.min, y.max = y.max, 
 .     y.trans = y.trans, y.nbreaks = y.nbreaks, sort = sort, stack = stack, 
 .     flip = flip, comparisons = comparisons, ref_group = ref_group, 
 .     pairwise_method = pairwise_method, multiplegroup_comparisons = multiplegroup_comparisons, 
 .     multiple_method = multiple_method, sig_label = sig_label, 
 .     sig_labelsize = sig_labelsize, aspect.ratio = aspect.ratio, 
 .     title = title, subtitle = subtitle, xlab = xlab, ylab = ylab, 
 .     legend.position = legend.position, legend.direction = legend.direction, 
 .     theme_use = theme_use, theme_args = theme_args, force = force, 
 .     seed = seed)
3. setNames(colnames(df_table)[apply(df_table, 1, function(x) which(x > 
 .     0))], rownames(df_table))

[hermespara]

I'm sorry. I cannot reproduce this error.
Maybe you can try to reproduce the df_table and check if there's something wrong.
df_table <- table([email protected][["celltype"]], [email protected][["celltype"]])
setNames(colnames(df_table)[apply(df_table, 1, function(x) which(x > 0))], rownames(df_table))