Because geometric mean ignores the positive predictive value (also called precision) (17), we report all values of the contingency matrix (Table 2), and summarize the predictive ability of the model with the Matthews correlation coefficient (18), a metric that is robust across a range of prevalence values and incorporates positive and negative predictive values.
As a summary metric, the accuracy of a prediction, Acc, can then be defined as the geometric average of sensitivity and positive predictive value, (8) As we gained more and more biological knowledge about the proteins, we can associate a protein with (possibly multiple) functional annotations.
(13) P r e c i s i o n = N c p | P |, R e c a l l = N c b | B | (14) F - M e a s u r e = 2 * P r e c i s i o n * R e c a l l P r e c i s i o n + R e c a l l In addition, sensitivity (Sn), positive predictive value (PPV) and geometric accuracy (Accuracy) have recently been proposed to evaluate the quality of protein complex predictions [ 7, 36, 42].
The accuracy measure is the geometric mean of two other measures: positive predictive value (PPV) and sensitivity.
The Yeast 5 GEM sensitivity, specificity, positive predictive value, negative predictive value, and geometric mean could all be improved by the reduction of false positive predictions - simulations which predict that biomass can be produced although reactions annotated as being catalyzed by "essential genes" have been blocked.
So it is necessary to balance the two measures by introducing the geometric accuracy (Acc), which is simply the geometric mean of the clustering-wise sensitivity and the positive predictive value: (5) A c c = S n × P P V The third measure we used is the maximum matching ratio (MMR) which was introduced in [ 6].
The second measure we used is the geometric accuracy as introduced by Broh´ee and van Helden [ 22], which is the geometric mean of two other measures, namely the clustering-wise sensitivity (Sn) and the clustering-wise positive predictive value (PPV).
In fact, earnings have no predictive value whatsoever.
The negative predictive value was 82%.
Negative predictive value was 99% and positive predictive value was 54%.
