Threshold-:
The threshold is used to
discriminate the cleavage and non-cleavage sites.The user can vary the
threshold score between the -1.5 to 1.5. The residues achieving score
more then the cutoff score are predicted as P1 cleavage site otherwise
they are predicted as non-cleavage site.if the user did not select
threshold then the default threshold of prediction methods will be
used.The default threshold is that at which the maximum MCC of
prediction method is achieved. The default threshold for in vitro digested data based and MHC ligands based classifiers is "-0.1" and "0.3" respectively. The higher the threshold (= high stringency), the lower the false positive rate and the hi
gher the false negative rate. in contrast the low the threshold (= low stringency), the higher the false positive rate and the lower the false negative rate. In short, from the same protein sequence input, a threshold setting of 1.0 will predict a lower
number of cleavage sites (cleavage sites with high score), compared to 0.0 or lesser thresholds.
Prediction Results-:
The results of the prediction displayed in
user-friendly text formats.Each of result display format firstly
provides a comprehensive account of length of input sequence, prediction
approach, selected classifier and cutoff threshold as shown below. The result
of prediction will be displayed in these two formats.
