Description Usage Arguments Value Examples
Idea is to shift co-ordinates so that the centroid of the negative control
is always centered on 0,0. Useful for PCA biplots where the 0,0 denotes the
centroid of the entire data, and not necessarily the negative control. If
given a dataframe and specified columns for the x and y co-ordinates, this
function will shift the co-ordinates to be centered on the centroid of a
specified compound. The difference between centre_control
and
centre_controls_m
is that centre_control_m
can centre a range
or columns, whereas centre_control
is limited to 2 named columns.
1 | centre_control(df, cols, cmpd_col, cmpd)
|
df |
dataframe |
cols |
column indicies or names |
cmpd_col |
column containing names to subset with |
cmpd |
name of compound to subset |
dataframe of 'df' shifted so that cmpd is centered on 0,0
1 2 3 4 5 6 7 8 | # artifical example using iris
pca <- prcomp(iris[,1:4])$x[,1:3]
df <- data.frame(pca, name = iris[,5])
pca_shift <- centre_control(df, 1:3, cmpd_col = 'name', cmpd = 'setosa')
par(mfrow = c(2,1))
plot(df$PC1, df$PC2, col = df$name, pch = 20)
plot(pca_shift$PC1, pca_shift$PC2, col = pca_shift$name, pch = 20)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.