Can anyone advise about what data to plot in the ROC curve to obtain the sensitivity and specificity of a mRNA as a disease biomarker?

Question

[tazyeen.safia]

Dear All: can anyone advise about what data to plot in the ROC curve to obtain the sensitivity and specificity of an mRNA as a disease biomarker? we are validating our target mRNAs comparing their level in healthy controls and patients. the data that we have after validation is either raw ct values, normalized ct values or fold changed. which one of those is better to be used to generate the ROC curve and can you explain why?

Answer 1

[Jean-Karim Heriche]

The ROC curve is a plot of the true positive rate (aka sensitivity) vs the false positive rate (aka 1-specificity) of a classifier. Therefore the data you need is a binary classifier output and some ground truth.