Open KristinaGagalova opened 6 months ago
Hi, Sorry for the late reply. We didn't define a high-quality prediction.
If the prediction score is less than 0.3, it is labeled as non-enzyme for Level 0 (enzyme, non-enzyme). If it is greater than one of the main classes' thresholds, it is labeled as that main class. Otherwise, it is predicted as no prediction.
There is a file called thresholds.txt under ECPred/lib/subclasses. It contains threshold values for each class and ECPred uses this file to give a positive prediction. If the prediction score is higher than the threshold for a given class, then it is a positive prediction.
So, the confidence score works like that.
Hi,
Could you please give me some more info about the confidence score? How would you define a high-quality prediction, for both enzymes and non-enzymes and how to select an HQ dataset?
Than you in advance