create ROC curve and determine optimal cut-off and establishing PPV

Using the below script I was able to create a labeled and coloured boxplot and calculate p-value for 2 classes, eg LB/B and LP/P which were established by VKGL.
Now I want to create an ROC curve for both classes thereby determining the optimal cut-off (youden index) and in addtion determine the PPV. Is possible to integrate this in the below script?

Does anybody have an idea how to do this?

Note, I have already installed following packagess





my_data %>%

  • filter(classification %in% c("LB/B", "LP/P") & Source %in% c("VKGL"))%>%
  • ggplot(mapping=aes(x=classification, y= Backbone Hbond, color="red", fill = "orange", alpha=0.2))+
  • geom_boxplot()+
  • labs(title = "Backbone Hbond for LB/B and LP/P classes (VKGL)",
  •    x = "Classification",
  •    y = "Backbone Hbond")+
  • stat_compare_means(label.x = 1.3, size = 4)



This topic was automatically closed 42 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.