In the present study, GoldScore-based classification correctly predicted 48 agonists and 120 nonagonists, whereas the hybrid scoring scheme correctly predicted 45 agonists and 137 nonagonists [Supplemental Material, Table 2 (doi:10.1289/ehp.1001930)].
In this analysis, we excluded cases for DO with RE and DO only in which the HSA was experiencing a stock-out of the drug needed to treat a classification correctly.
Leave-one-out-classification correctly classified 40% of Group 1, 44.4% of Group 2, 73.3% of Group 3, and 82.2% of Group 4, with Groups 1 and 2 being misclassified as each other 40% of the time.
As the C-statistic is often criticised for its limited clinical interpretability (Vickers, 2011), we have further provided oncologists with the practical approach of using NRI to gauge the extent to which the new prognostic classification correctly reclassifies patients' levels of risk.
51/109(46.8%) children with severe classifications were correctly identified.
In addition, all 12 short-peptide PPIs that were omitted from the training set as trivial classifications are correctly identified (not shown).
In other words, it classifies how well a binary classification test correctly identifies a condition.
Taking the neutral apatite [La8Sr2][(GeO4)6]O2 as a prototype electrolyte, this classification scheme correctly predicted that 'excess' oxygen in La9SrGe6O26.5 is tenanted in the framework as [La9Sr][(GeO4 5.5(GeO5 0.5]O2, rather than the presumptive tunnel location of [La9Sr][(GeO4 6]O2.5.
Classification accuracy was determined also for the traditional CNS PET tracer selection criteria (Figure 1) and illustrated in Figure 4. Second to the presented ftb classification, which correctly classified all of the non-functioning tracers, was the Vu,brain criterion (Vu,brain ≤20 mLg brain-1) resulting in 10/12 correct classifications.
Using the docking consensus classification, we correctly classified 15 of 28 (55.6%) molecules from this sample ToxCast data set.
