False positive rate

A new tumor marker is evaluated using the same criteria used for many diagnostic tests (i.e., sensitivity, specificity, and accuracy). The diagnostic sensitivity and specificity are best represented by a receiver operating characteristic (ROC) curve. The ROC curve is constructed with the true-positive rate versus false-positive rate at various decision levels. As a test improves in its diagnostic performance, it shifts upward and to the left as the true-positive rate increases and the false-positive rate decreases. 

False-positive results with bDNA have been observed with proficiency testing specimens for HTV-1 in the College of American Pathologists HIV-1 viral load survey and HCV in the viral quality control program administered by the Netherlands Red Cross. The reason for the false-positive results with these proficiency testing specimens is not known but may be sample matrix effects. The extent to which this problem occurs with clinical samples has not been determined. However, both the HIV-1 and HCV bDNA assays were designed to have a false-positive rate of 5%. 

False positive rate probability that the test result is positive when the analyte is not present, A... 

PPR is the probability of obtaining positive responses, TPR true positive rate (Eq. 4.48), TNR true negative rate (Eq. 4.49), FPR false positive rate (Eq. 4.50), FNR false negative rate (Eq. 4.51), X/ and xu are the lower and upper limits of the unreliability region... 

Finally, one of the main limitations of this model is the large false-positive rate and large experimental errors. Indeed, the primary limitation of the QSAR and in silico models is the high false-positive rates in oral bioavailability predictions. 

These are well-known classification parameters true positive rate (p ), false positive rate (pnx), true negative rate (qnx), and false negative rate (qK). They can be easily obtained from the previous computations where we calculated the number of taxon and nontaxon members in each interval. For example, to calculate the true positive rate, we sum the number of taxon members in intervals above the hitmax, plus half of taxon members in the hitmax interval and divide this by the total number of taxon members in the sample. To calculate the false negative rate, we sum number of taxon members in intervals below the hitmax, plus half of the taxon members in the hitmax interval and divide this by the total number of taxon members in the sample. 

As a result of this protocol, four indicators were dropped because in each case, they did not pass the first consistency test, that is, failed to discriminate adequately at all levels of the scale. Next, Tyrka et al. (1995) calculated the taxon base rate for each indicator using a hybrid of MAXCOV and Latent Class Analysis estimation procedures (for details see Golden, 1982) and adjusted the estimate for the true- and false-positive rates computed earlier. The average taxon base rate was. 49. The authors did not report a variability statistic, but a simple computation shows that SD of base rate estimates was. 04. 

Several algorithms are available for the analysis of MS/MS spectra including SEQUEST, MASCOT, and X Tandem among others. Note that additional secondary quality control of assessment of MS/MS data has recently been implemented to assess identification probabilities and false positivity rates. The MS/MS spectra from an experiment can be interrogated against a concatenated forward and reverse database and an assessment of the intrinsic error rate of the data set can be made. Other approaches for secondary analysis of matching scores for peptide sequencing data include XCorr score normalization routines that are independent of peptide and database size.33... 

Greater resources required Less useful conclusions Stronger assumptions required Higher statistical false-positive rate Blind to effects in the opposite direction... 

Another approach to controlling the false positive rate in carcinogenicity studies was proposed by Haseman (1983). Under this rule, a compound would be declared a carcinogen if it produced an increase significant at the 1% level in a common tumor or an increase significant at the 5% level in a rear tumor. A rare neoplasm was defined as a neoplasm that occurred with a frequency of less than 1% in control animals. The overall false positive rate associated with this decision rule was found to be not more that 7-8%, based on control tumor incidences from NTP studies in rats and mice. This false positive rate compares favorably with the expected rate of 5%, which is the probability at which one would erroneously... 

Haseman, J.K. (1983). A reexamination of false-positive rates for carcinogenicity studies. Fundam. Appl. Toxicol. 3 334—339. 

This is obviously an important determinant of the precision of the findings. The calculation of the appropriate number depends on (1) the critical difference, that is, the size of the effect it is desired to detect (2) the false positive rate, that is, the... 

The Scheffe procedure is powerful because of it robustness, yet it is very conservative. Type I error (the false positive rate) is held constant at the selected test level for each comparison. 

Because of the significant false positive rate for the ELISA test, a second, more specific test for HIV antibodies is also used the Western blot test. This technique has a lower incidence of false positives than the ELISA assay. In practice, serum samples that score antibody positive by the ELISA test are generally retested by the Western blot procedure. Serum samples are considered positive if they are found to contain HIV-specific antibodies by both tests. 

An alternative measure of reliability often used is the "false positive" rate, defined as ... 

The confirmation rate for the selected set was dramatically higher than the typical 60 to 80% observed historically, internally, and in the literature [30]. The results showed a 90.5% confirmation rate (9.5% false positive rate) at >30% inhibition, giving 3683 confirmed actives. The confirmation rate rose to 97.4% when considering only compounds selected from enriched clusters and fell to 84.9% when considering compounds selected by diversity. This supports the hypothesis that actives selected from highly enriched clusters are significantly more likely to confirm, that is, be true positives, than other compounds because of the presence of surrogate replicates . 

The results showed a 59% confirmation rate (41% false positive rate), giving 4296 confirmed actives. Of the compounds initially selected with >40% inhibition, 69% confirmed at >40% inhibition. This can be broken down further into compounds that feU into enriched clusters and those that did not, with confirmation rates of 73 and 68% respectively. In this case, there was only a modest increase in the confirmation rate for compounds selected from enriched clusters. This might be attributed to the fact that the enrichment within the active clusters was only about 50% greater than the background hit rate (21 versus 14.7%) in contrast to the peptide hydrolase case where the enrichment was 2.6-fold higher. 

The results showed a 46.6% confirmation rate (53.4% false positive rate), giving 1555 confirmed actives (1,096 unique ring hashcodes, 47.9%). A second round of filtering was applied using tighter physiochemical property criteria. Molecules with 200 > MW > 500, clogP > 5.0, rotors > 8 were flagged and visually assessed. The final selection for secondary assays was 1087 compounds (70%). 

Table IV gives values of J when the false positive rate is 5 percent and the false negative rate is either 5 percent or 1 percent. J depends on V and, since K = 1, on the sura a + a . A...

Table IV. Sample Size (Value of J Given K = 1) Required CO Ensure a False Positive Rate of 5 Percent and a False Negative Rate of Either 5 Percent or 1 Percent When Comparing Two Means...

Wang, X., Hessner, M.J., Wu, Y, Pati, N., and Ghosh, S., Quantitative quality control in microarray experiments and the application in data filtering, normalization and false positive rate prediction, Bioinformatics, 19, 1341-1347, 2003. 

False positive rate (%) = false positives 100/total known negatives... 

---