@schuemie ,
In comparison study skeleton, generateDiagnostics function
if (sum(!is.na(controlSubset$seLogRr)) >= 5) { fileName <- file.path(diagnosticsFolder, paste0("trueAndObs_a", analysisId, "_t", targetId, "_c", comparatorId, "_", label, ".png")) #controlSubset doesn't have targetEffectSize try( EmpiricalCalibration::plotTrueAndObserved(logRr = controlSubset$logRr, seLogRr = controlSubset$seLogRr, trueLogRr = log(controlSubset$targetEffectSize), fileName = fileName) ) }
In the controlSubset doesn't have targetEffectSize, so EmpiricalCalibration::plotTrueAndObserved doesn't work