How can I create a mean ROC curve for multiple plotted ROC curves, R?

Question

[sap940]

I have a dataset of 100 samples, each of which has 195 mutations with their corresponding known clinical significance ("RealClass") and predicted value according to some prediction tool ("PredictionValues")

For the demonstration, this is a random dataset that has the same structure as my dataset:

predictions_100_samples<-as.data.frame(matrix(nrow=19500,ncol=3))
colnames(predictions_100_samples)<-c("Sample","PredictionValues","RealClass")
predictions_100_samples$Sample<-rep(c(1:100), each = 195)
predictions_100_samples$PredictionValues<-sample(seq(0,1,length.out=19500))
predictions_100_samples$RealClass<-rep(c("pathogenic","benign"),each=10)
colours_for_ROC_curves<-rainbow(n=100)

I plotted all of those 100 sample as ROC curves via PROC package:

    library("pROC")
    roc_both <- plot(roc(predictor=predictions_100_samples[1:195,2],response = predictions_100_samples[1:195,3]), col = colours_for_ROC_curves[1],main="100 samples ROC curves",legacy.axes=TRUE,lwd=1)
    i=2
    for(i in 1:100){
        set.seed(500)
        roc_both <- plot(roc(predictor=predictions_100_samples[(((i-1)*195)+1):(i*195),2],response = predictions_100_samples[(((i-1)*195)+1):(i*195),3]), col = colours_for_ROC_curves[i], add = TRUE,lwd=1)
                         i=i+1
    }

And that is how the final plot looks like: - https://ibb.co/heCkxU

Now, I want to add the mean ROC curve of all 100 plotted ROC curves to the same plot. I tried to use the sensitivities and specificities calculated for each threshold via "roc" function along the loop I wrote (It can be achived by roc_both$sensitivities, roc_both$specificities, roc_both$thresholds)

But the main problem was that the chosen thresholds were random and not equal along the 100 ROC curves I plotted, so I could'nt calculate the mean ROC curve manually.

Is there a different package that may allow me to produce the mean ROC curves of multiple ROC curves? Or is there a package that allows setting the thresholds for calculating sensitivity and specificity manually, so I could later on be able to calculate the mean ROC curve? Do you maybe have a different solution for my problem?

Thank you !

[sap940]

[Ram]

Do not add answers unless you're answering the top level question. Edit your question and add the content there. Also, see: How to add images to a Biostars post

[Kevin Blighe]

Cross-posted: https://support.bioconductor.org/p/113334/

[sap940]

Dear Kevin, Does bioconductor and biostars run together in cooporation and host the exact same coders for my post to be called "Cross-posted"?

Anyways, a few days ago I posted this question at StacksOverflow, and noone knew the answer. I also asked several co-workers and noone knew the answer.

So a few hours ago I posted the same question here and in bioconductor as my last and desparate resort, because Im completely clueless. Was it against the rules ?

[Kevin Blighe]

There are both different and the same users on both sites. By mentioning the cross-posting, it is more to just alert everyone so that, if an answer is given in one place, people are not wasting their efforts. I can see, for example, that you now have an answer on Bioconductor from a user who is also active here.

My answer would be a question in return: why do you even want to obtain the average? Are these simply 100 bootstraps or cross-validations of the same data? If so, just run the model through cv.glm or cv.lm and obtain the change in delta to infer robustness of the model. If delta is small, model can be assumed to be robust (although many other factors go into this).

I go over some of this here: A: Resources for gene signature creation

You may not have obtained an answer on StackOverflow because many would immediately frown at the thought of obtaining an average of multiple ROC curves.