Differential analysis & tests / ROC curve
Your data stays in your browser: files are read and processed on your device, never uploaded to any server, and cleared when you close the page. Free, no sign-in.
A ROC curve shows sensitivity (true positive rate) against 1 − specificity (false positive rate) over all cut-offs; curves nearer the top-left discriminate better. The dashed diagonal is chance (AUC = 0.5).
The AUC is the probability that a random positive scores higher than a random negative; the table gives its DeLong standard error and confidence interval.
The Youden cut-off maximises sensitivity + specificity − 1. Treat it as a guide: clinical cut-offs often weigh missed and false diagnoses differently, and choosing and evaluating a cut-off on the same data is optimistic, so validate it in independent samples.
When several markers are measured on the same samples, their AUCs are compared with the paired DeLong test (samples with both values only), with P values adjusted by the method you choose.
With direction “Auto”, a marker whose AUC is below 0.5 is scored as “lower = positive”, and the table says so.
DeLong, DeLong & Clarke-Pearson (1988) Biometrics 44:837; Sun & Xu (2014) IEEE Signal Process Lett 21:1389 (fast algorithm); Youden (1950) Cancer 3:32. Curve coordinates and AUC match sklearn.metrics.roc_curve / roc_auc_score.
Up to 200,000 rows.
Send us your data and research question and you will receive a written plan within 1 working day: analysis steps, parameter rationale, deliverables and timeline. Quoted per project.