At a classification threshold of 0.75 for present connections and 0.25 for absent connections, we made predictions about the existence of 418 unknown connections.
For a classification threshold of 500 nM, we find that the method has a specificity of 0.95, and a sensitivity of 0.74.
A classification threshold is then determined for each network that maximizes Youden's index J for the training set: J = sensitivity + specificity – 1 (9).
One major limitation of prior computational methods used for this purpose is the use of a universal classification threshold for all genes at all taxonomic levels.
One major limitation of prior methods used in this context is the use of a universal classification threshold for all genes at all taxonomic levels (e.g., BLASTP E-value=0.1 used by AMPHORA [ 12]).
In practical terms this might apply to the classification of traits such as obesity in humans, based on height and weight, or schizophrenia based on a cumulative score for a set questionnaire, where a classification threshold needs to be crossed for an individual to be classified as affected.
Figure 17 shows the distributions of the opening-ratios as a function of the classification threshold for each of the relevant structures.
We applied a LOOCV loop to determine max AUC,1-AUC) and optimax AUC,1-AUCandoptimalshold for the signature metagene in eaclassification
The central box highlights tallies that straddle the classification threshold for the sum. Figure 10 shows the results of applying our beta binomial analysis to the training pool for model logP3-3a.
