CheyennedeVriend
commented
2 years ago Hi there, I ran into the same problem previously. I found the support function online. You can copy paste the following piece of code:
#'\code{magicPlotMakerLight}
#'
#'This function is designed to generate and save a battery of plots aimed at helping visualise a measurments dataframe.
#'The function generate 3 plots to directly visualise measurements in the form of a cloud of points, a series of boxplots and surperimposed density curves.
#'It also genereate a pca plot (using nicePCA unction if a targets table is provided).
#'Finally it generates two heatmap, on directly clustering over the measurments and another one performing a clustring over the cross-correlation matrix of the samples.
#'The plots are directly saved to a folder. Nothing is returned.
#'
#'@param df the measurment n*m dataframe (n is number of omic features, m is number of samples) where columns are ordered by conditions.
#'@param targets A n*2 dataframe, where n is the number of samples. First column correspond to samples, second column correspond to conditions.
magicPlotMakerLight <- function(df, targets = NULL, no_pca_label = FALSE)
{
if (!is.null(targets))
{
df_and_targets <- make_df_and_targets_great_again(df,targets)
df <- df_and_targets[[1]]
targets <- df_and_targets[[2]]
}
##This part is just to generate the melted dataframe for ggplot
df$ID <- row.names(df)
melted_df <- reshape::melt(df)
index <- c(1:length(melted_df[,1]))
df <- df[,-length(df[1,])]
#########################################
## ggplots ##
#########################################
violins <- ggplot(melted_df, aes(x = index, y = value, group = variable, color = variable)) + geom_violin() + theme_minimal()
# plot(violins)
complete_df <- df[complete.cases(df),]
t_complete_df <- t(complete_df)
if (is.null(targets)){
PCA <- prcomp(t_complete_df ,center = TRUE, scale. = T)
pcaPlot <- plot(PCA$x[,1],PCA$x[,2], pch = 19, xlab = paste("PC1 (",as.character(round(PCA$sdev[1]^2/sum(PCA$sdev^2)*100)),"%)"), ylab = paste("PC2 (",as.character(round(PCA$sdev[2]^2/sum(PCA$sdev^2)*100)),"%)"), main = "PCA of samples")
text(PCA$x[,1],PCA$x[,2], labels = names(PCA$x[,1]), pos = 3)
}
else
{
if (no_pca_label)
{
pcaPlot <- nicePCA(df = complete_df, targets, c(1,2,3), no_label = T)
}
else
{
pcaPlot <- nicePCA(df = complete_df, targets, c(1,2,3))
}
}
# plot(pcaPlot)
return(list(violins,pcaPlot))
}
plotTF <- function(TF, geneLevelStats, regulon_df)
{
names(geneLevelStats) <- c("ID","stat")
targets <- regulon_df[regulon_df$TF == TF,]
names(targets) <- c("ID","TF","sign")
target_stats <- merge(geneLevelStats,targets, by = "ID")
target_stats$stat <- target_stats$stat * target_stats$sign
target_stats <- target_stats[order(target_stats$stat, decreasing = T),]
target_stats <- unique(target_stats)
target_stats$ID <- factor(target_stats$ID, levels = target_stats$ID)
plot(ggplot(target_stats,aes(x = ID, y = stat, fill = stat))+ geom_bar(stat = "identity"))
}Hope it helps.
Kind regards, Cheyenne
dilipgosar
gabora
Dear Sir/ Madam,
I am trying R code for normalisation in transcriptutorial. I am getting following error at this step
It seems it is not able to find magicPlotMakeLight function. I like to know to which package this function belongs and if you can fix this code.
Appreciate your help.
Sincerely,
Dilip Gosar