False discovery rate

This also serves to highlight the search specificity in that an increasing false discovery rate would have an adverse effect on correlation and confidence measures. 

Abbreviation Bp, nucleotide base pairs cDNA, complementary DNA ChIP, chromatin Immunoprecipi-tation Cy5, cyanine 5-dCTP Cy3, cyanine 3-dCTP ESTs, expressed sequence tags FDR, false discovery rate MIAME, minimum information about a microarray experiment mRNA, RNA, messenger NIA, National Institutes of Aging RFUs, relative fluorescence units RT-PCR, reverse transcriptase polymerase chain reaction SAGE, serial analysis of gene expression SAM, significance analysis of microarrays... 

Yang Q, Cui J, Chazaro I, Guppies LA, Demissie S. (2005) Power and type I error rate of false discovery rate approaches in genome-wide association studies. BMC Genet. 6(suppl 1), SI34. 

Statistical analysis should be appropriate to the types of outcome data collected and the number of genotypes nsed in the analysis. The handling of missing data should be clearly stated. Corrections for multiple comparisons (e.g., controlling for false discovery rates 36) should be performed if multiple statistical tests are carried out. 

Benjamini, Y., and Hochberg, Y. (1995) Controlling false discovery rate a practical and powerful approach to multiple testing. J R Statist Soc Br. 57, 289-300. 

Liu P, Hwang JT (2007) Quick calculation for sample size while controlling false discovery rate with application to microarray analysis. 

Demrrkale CY, Nettleton D, Maiti T (2010) Linear mixed model selection for false discovery rate control in microarray data analysis. Biometrics 66 621-629. doi BIOM1286 (pii)10.1111/j.l541-0420.2009.01286.x... 

Efron B, Tibshirani R. Empirical bayes methods and false discovery rates for microarrays. Genet Epidemiol 2002 23 70-86. 

SAM involves a modified t-test and computes a False Discovery Rate (FDR, representing the expected incidence of false positives) for each chosen differential expression (significance) cut-off. 

The complementary rates are obtained from the opposite conditioning the fraction of model-predicted hits that are observed hits (A A) and the fraction of model-predicted inactives (I /) that are observed inactives. We call these the positive discovery rate and negative discovery rate . It is important to look at these conditional probabilities a very clear example is in the analysis of gene chip microarray data where the false discovery rate is 1 minus the positive discovery rate as defined above and in Chapter 6 an excellent discussion is given by Benjamini and Hochberg (1995). 

Belief that a very small p-value for a predictor (for example, a biomarker) is more likely to occur with high predictive accuracy. The multiple testing problem must not be ignored, and the false discovery rate (FDR) controlled see also Chapter 6. 

Suppose that we wish to make inferences on the parameters 0i,i = 1,g, where 9i represents the logarithm of the ratio of the expression levels of gene i under normal and disease conditions. If the ith gene has no differential expression, then the ratio is 1 and hence 0 = 0. In testing the g hypotheses Ho, 0 = 0, / = 1,..., g, suppose we set R, = 1 if H0, is rejected and Ri = 0 otherwise. Then, for any multiple testing procedure, one could in theory provide a complete description of the joint distribution of the indicator variables R, ..., Rg as a function of 0i,..., 0g in the entire parameter space. This is impractical if g > 2. Different controls of the error rate control different aspects of this joint distribution, with the most popular being weak control of the familywise error rate (FWER), strong control of the familywise error rate, and control of the false discovery rate (FDR). 

Efron, B., Tibshirani, R., Storey, J. D., and Tusher, V. (2001). Empirical Bayes analysis of a microarray experiment. Journal of the American Statistical Association, 96, 1151-1161. Firmer, H. and Roter, M. (2001). On the false discovery rate and expected type I errors. Biometrical Journal, 43, 985-1005. 

FDR False discovery rate, the expected proportion of rejected null hypotheses that are false positives. 

Tong T, Zhao H. Practical guidelines for assessing power and false discovery rate for a fixed sample size in microarray experiments. StatMed. 2008 27 1960-1972. 

Y. Benjamini and Y. Hochberg, Controlhng the false discovery rate A practical and powerful approach to multiple testing. StatSoc SerB (Methodological) 57 289-300... 

Y. Benjamini and D. Yekutieh, The control of the false discovery rate in multiple tests under dependency. Ann Stat 29 1165-1188 (2001). 

J. D. Storey, A direct approach to false discovery rates. J Roy Stat Soc Ser B 64 479-498... 

---