【論】Storey,2003,Statistical significance for ～

John D.Storey and Robert Tibshirani
Statistical significance for genomewide studies
Proc Natl Acad Sci USA 2003, 100:9440-9445.3.2
[PDFダウンロード][Webサイト]

・DNAデータの統計処理にq値の使用を提案する。
・q値の計算法や、p値との比較が主な内容。
・使用データ
1.Detecting Differentially Expressed Genes.[Hedenfalk]
2.Identifying Exonic Splicing Enhancers.[Fairbrother]
3.Genetic Dissection of Transcriptional Regulation.[Brem]
4.Finding Binding Sites of Transcriptional Regulators.[Lee]
・q値を計算するソフト→[ダウンロード]

・概要「We propose that the recently introduced q value is a well suited measure of significance for this growing class of a genomewide tests of signifcance. The q value is an extension of a quantity called the "false discovery rate"(FDR),」
・「Whereas the p value is a measure of significance in terms of the false positive rate, the q value is a measure in terms of the FDR.」
・「For example, a false positive rate of 5% means that on average 5% of the truly null features in the study will be called significant. A FDR of 5% means that among all features called siginificant, 5% of these are truly null on average.」
・「For example, controllong the FDR at 0.03, 0.05, or 0.07 in the expression data finds 80, 160, or 231 significant genes, respectively, when our proposed method is used. The methodology in ref.5 finds only 21, 88, or 153, respectively, indicating that this eariler method's estimates are too conservative and result in a substantial loss of power.」
・「Most inportantly, a systemaic use of q values in genomewide tests of significance will yield a clear balance of false positives to true positive results and give a standard measure of significance that can be universally interpreted.」
・q値の定義「An alternative quantity, called the pFDR, was recently proposed, which is simply defined as pFDR = E[F/S|S>0]. The q value most technically defined as the minimum pFDR at which the feature can be called significant.」

・"q value"のイメージがサッパリわかない。もうちょっといろいろ調べないと歯が立たない。
・とても広い範囲を含む題目。よっぽど自信があるんだろうなぁ。。。