【論】Kerr,2001,Bootstrapping cluster analysis: ～

M.Kathleen Kerr and Gary A.Churchill.
Bootstrapping cluster analysis: Assessing the reliability of conclusions from microarray experiments.
Proc Natl Acad Sci USA. 2001 Jul 31;98(16):8961-5. Epub 2001 Jul 24.
[PDF][Web Site]

・Bootstrapping cluster analysisのアレイデータ解析への応用
・データ：Yeast、6118遺伝子、時間点7点、cDNA[Chu]

・概要「Here we apply bootstrapping to assess the stability of results from a cluster analysis.」
・目的「The goal of bootstrap clustering is to statistical inference about a discrete structure,」

・"bootstrap"の原理がわからない。擬似データを作ってデータを水増しして、解析精度を高めるということ？？
参考：http://www.is.titech.ac.jp/~shimo/msj2004/node2.html　グッハァァ・・・orz

【論】Okada,2005,Knowledge-assisted recognition of ～

Y.Okada, T.Sahara, H.Mitsubayashi, S.Ohgiya, T.Nagashima.
Knowledge-assisted recognition of cluster boundaries in gene expression data.
Artif Intell Med. 2005 Sep-Oct;35(1-2):171-83.
[PDF]

・遺伝子データベースの情報を利用して、クラスタの閾値を自動的に決定する方法を提案する。
・データ：その1 Yeast, Cell cycle data, 416遺伝子[Cho] その2 Yeast, Cold stress response data, 1609遺伝子[Sahara]
・クラス分け法：Hierarchical clustering, K-means, SOM, AutoClass
・クラス分け評価指標：The rand index, The cluster separation, The cluster entropy

・目的「The aim of our algorithm is to find cluster boundaries so as to compose functionally independent agglomerations on the dendrogram derived from similarities among gene expression patterns.」
・アルゴリズム「The algorithm proposed in this paper is composed of three steps: (1) hierarchical clustering based on gene expression profiles, (2) automatic determination of the cluster boundaries using the VIF among the Gene Function Vectors representing distributions of gene functions in each cluster, and (3) annotation for each cluster according to dominant gene functions within the respective clusters.」

《チェック論文》
[1]Sahara T, Goda T, Ohgiya S.,Comprehensive expression analysis of time-dependent genetic responses in yeast cells to low temperature., J Biol Chem. 2002 Dec 20;277(51):50015-21. Epub 2002 Oct 11.
[2]Cho RJ, Campbell MJ, Winzeler EA, Steinmetz L, Conway A, Wodicka L, Wolfsberg TG, Gabrielian AE, Landsman D, Lockhart DJ, Davis RW.,A genome-wide transcriptional analysis of the mitotic cell cycle., Mol Cell. 1998 Jul;2(1):65-73.
[3]Yeung KY, Fraley C, Murua A, Raftery AE, Ruzzo WL.,Model-based clustering and data transformations for gene expression data.,Bioinformatics. 2001 Oct;17(10):977-87.
～～～～～～～～～～
・論文記録が50本目ということで、今回はスペシャルです。どの辺がスペシャルかは、わかる人だけわかります。ツッコミ不要です。
・英英辞書を使うつもりが、どうしても英和辞典に戻ってしまいます。専門用語は英和の方がずっと充実しているし。

【論】Dudoit,2002,Statistical methods for identifyin～

Sandrine Dudoit, Yee Hwa Yang, Matthew J.Callow and Terence P.Speed.
Statistical methods for identifying differentially expressed genes in replicated cDNA microarray experiments.
Statistica Sinica 12(2002), 111-139
[PDF][Web Site]

・マイクロアレイデータの統計解析の概観とその応用。
・データ：マウス。正常(control)×8、apo AI gene knocked-out×8、合計16サンプル。257遺伝子。
・遺伝子抽出法：adjusted p-values

・「Given suitably normalized data, the biological question of differential expression is restated as a problem in multiple hypothesis testing: the simultaneous test for each gene of the null hypothesis of no association between the expression levels and the treatment/control status.」
・手順「Our basic approach for identifying differentially expressed genes consists of two steps: (1) computing a test statistic for each gene, and (2) obtaining adjusted p-values for a multiple testing procedure which strongly controls the family-wise Type I error rate and takes into account the dependence structure between the gene expression levels.」
・「To date, attention has focused primarily on expression at the transcription stage, i.e., on mRNA or transcript levels.」
・目的「The goal of the study is to identify genes with altered expression in the livers of two lines of mice with very low HDL cholesterol levels compared to inbred control mice.」
・注意点「Note that the spotted cDNA sequences are usually referred to as "genes", whether they are actual genes, ESTs (expressed sequence tags), or DNA sequences from other sources.」

・内容がつかめない。図（M-A plot）の見方がわからず。

《チェック論文》
・Chen, Y., Dougherty, E. R. and Bittner, M. L. (1997). Ratio-based decisions and the quantitative analysis of cDNA microarray images. J. Biomedical Optics 2, 364-374.
・Kerr, M. K. and Churchill, G. A. (2001). Experimental design for gene expression microarrays. Biostatistics 2, 183-201.

【論】Efron,2001,Empirical Bayes analysis of a ～

Bradley Efron, Robert Tibshirani, John D.Storey, and Virginia Tusher.
Empirical Bayes Analysis of a Microarray Experiment.
Journal of the American Statistical Association, 96: 1151-1160.
[PDF][Web Site]

・ベイズ統計を使ってデータの次元を下げる手法について。
・データ：ヒト、Lymphoblastoid cell lines GM14660 and GM08925、Affy.、6810遺伝子、8サンプル

・「The empirical Bayes inferences are closely related to the frequentist false discovery rate (FDR) criterion.」
・目的「One goal in this article is to provide a method for comparing the statistical efficiency of different data reduction strategies.」

・ベイズ統計とFDRの何たるかがわかっていないので、サッパリ内容つかめず。

《チェック論文》
・Kerr MK, Churchill GA.,Bootstrapping cluster analysis: assessing the reliability of conclusions from microarray experiments., Proc Natl Acad Sci USA. 2001 Jul 31;98(16):8961-5. Epub 2001 Jul 24.

【論】Storey,2003,Invited comment on 'Resampling- ～

John D.Storey
Invited comment on 'Resampling-based multiple testing for DNA microarray data analysis' by Ge, Dudoit, and Speed.
Test, 12: 1-77.
[PDF][Web Site]

・Dudoit, Speed等による論文に対するコメント。元の論文を読んでいないので、話がさっぱり見えません。文献の選択ミス。　それ以前に数学の力不足。生物の話もわからない。数学の話もわからない。でいったいどうするんでしょうか。マイッタ。

【論】Alizadeh,2000,Distinct types of diffuse large ～

Ash A.Alizadeh, M.B.Eisen, R.E.Davis, C.Ma, I.S.Lossos, A.Rosenwald, J.C.Boldrick, H.Sabet, T.Tran, X.Yu, J.I.Powell, L.Yang, G.E.Marti, T.Moore, J.Hudson Jr, L.Lu, D.B.Lewis, R.Tibshirani, G.Sherlock, W.C.Chan, T.C.Greiner, D.D.Weisenburger, J.O.Armitage, R.Warnke, R.Levy, W.Wilson, M.R.Grever, J.C.Byrd, D.Botstein, P.O.Brown, and L.M.Staudt.
Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling.
Nature, 403:503-511, 2000.
[PDF][Web Site]

・マイクロアレイデータによる病気診断に関する初期の論文。従来ではできなかった診断が、マイクロアレイを使うと可能になるという主旨。
・データ：Diffuse large B-cell lymphoma（びまん性大細胞型リンパ腫）、アレイ数128、サンプル数96（96名からの128組織？）。
・クラス分け法：Hierarchical clustering [Eisen]

・概要「We addressed whether we could (1) generate a molecular portrait of distinct types of B cell malignancy; (2) identify distinct types of B-cell malignancy not recognized by the current classification system; and (3) relate each malignancy to normal stages in B-cell development and physiology.」
・展望「By profiling the gene expression of many more DLBCLs, it may become possible to implicate a single gene or pathway in chemotherapy responsiveness with statistical certainty.」

【論】Arava,2003,Genome-wide analysis of mRNA ～

Yoav Arava, Yulei Wang, John D.Storey, Chih Long Liu, Patrick O.Brown, and Daniel Herschlag
Genome-wide analysis of mRNA translation profiles in Saccharomyces cerevisiae
Proc Natl Acad Sci USA. 2003 April 1;100(7):3889-3894
[PDF][Web Site]

・数あるマイクロアレイ解析のうち、酵母のmRNAに焦点をあてたものだが、その研究背景の時点で理解できず。理解度5%・・・　orz

・展望「One of the goals of global translational analysis was to provide a clearer quantitative picture of the link between mRNA and protein levels.」
・展望「Future work is needed to assesss and improve the quantitative accuracy of microarray data to determine the extent and character of variation in elongation and termination for different mRNAs and to provide direct quantitative measures of protein synthesis for large numbers of proteins.」

【論】Dudoit,2002,Comparison of discrimination ～

Sandrine Dudoit, Jane Fridlyand, Terence P.Speed
Comparison of discrimination methods for the classification of tumors using gene expression data
J.Am.Stat.Assoc.97 (2002) 77.
[PDF][Web Site]

・ヒトのマイクロアレイデータを使って、解析法の性能比較をする。
・データ
1.Leukemia (ALL/AML) [Golub]
2.Lymphoma [Alizadeh]
3.60 cancer cell line (NCI60) [Ross]
・前処理：BSS/WSS ratios の値により解析に使う遺伝子を抽出する
・解析法
1.Discriminant analysis : FLDA, DLDA, DQDA, Golub
2.CART-based classifiers : CART, Bag, CPD, MVN, Boost
3.Nearest neighbor classifiers
・評価法: 2:1 schemeでクラス識別をし、その error rate で評価する

・「In the words of Alizadeh et al.:"Indeed, the new methods of gene expression profiling call for a revised definition of what is deemed a 'disease'."」
・「The rankings of the classifiers were similar across datasets and the main conclusion, for these datasets, is that simple classifiers such as DLDA and nearest neighbors perform remarkably well compared to more sophisticated methods such as aggregated classification trees.」

《チェック論文》
・A.A.Alizadeh, M.B.Eisen, R.E.Davis, C.Ma, I.S.Lossos, A.Rosenwald, J.C.Boldrick, H.Sabet, T.Tran, X.Yu, J.I.Powell, L.Yang, G.E.Marti, T.Moore, J.Hudson Jr, L.Lu, D.B.Lewis, R.Tibshirani, G.Sherlock, W.C.Chan, T.C.Greiner, D.D.Weisenburger, J.O.Armitage, R.Warnke, R.Levy, W.Wilson, M.R.Grever, J.C.Byrd, D.Botstein, P.O.Brown, and L.M.Staudt. Difierent types of difiuse large b-cell lymphoma identified by gene expression profiling. Nature, 403:503-511, 2000.

【論】Simon,2003,Pitfalls in the use of DNA micro～

Richard Simon, Michael D.Radmacher, Kevin Dobbin, Lisa M.McShane
Pitfalls in the use of DNA microarray data for diagnostic and prognostic classification
Natl.Cancer Inst.95 (2003) 14.
[PDFダウンロード]

●目次
・Experimental objectives drive design and analysis
・Limitations of cluster analysis for class prediction
・Class prediction using supervised methods
・Cross-validation of prediction accuracy
・Validation on independent data
・Reporting the error rate
・Claims for new class prediction methods
・Conclusion

・注意点「Although cluster analysis appropriate for class discovery, it is often not effective for class comparison or class prediction.」
・問題点「Cluster analysis also does not provide statistically valid quantitative information about which genes are differentially expressed between classes.」
・問題点「One major limitation of supervised methods is overfitting the predictor.」
・結論「We recommend that supervised methods rather than cluster anlyses be used for class prediction and class comparison studies.」
・結論「If cross-validation is used to estimate prediction accuracy, then the entire model-building process, including the selection of informative genes, should be repeated in each cross-validation training set.」

【論】Tsai,2005,Multi-class clustering and predict ～

Chen-An Tsai, Te-Chang Lee, I-Ching Ho, Ueng-Cheng Yang, Chun-Houh Chen, James J.Chen
Multi-class clustering and prediction in the analysis of microarray data
Math Biosci. 2005 Jan ;193:79-100
[PDFダウンロード]

・8種類の物質（金属）を与えたアレイデータの網羅的な解析をし、general frameworkを確立する。
・データ：ヒト（skin fibroblasts）、cDNA、コントロール＋金属8種（As,Cd,Ni,Cr,Sb,Pb,Cu,AsV）合計55サンプル。
・実験手順
1.Significance testing for gene selection : variance F-testにより遺伝子を抽出し、5つのセットに分類する
2.Cluster analysis　: Hierarchical clustering and modified partition (k-means) methods
3.Class prediction : Fisher's linear discriminant analysis and k-nearest neighbors algorithms

・これまでの結果「They concluded that simple classification methods such as linear discriminant and nearest neighbors methods tend to perform well compared to other more sophisticated methods.」
・問題点「One problem with the k-means method is that the result often depends on the initial points. Different starting points may result in different clustering partitions. That is, the k-means algorithm may often find the local optimum.」

・あまりに内容豊富で焦点がハッキリせず、内容が把握しずらい。

《チェック論文》
・S.Dudoit, J. Fridlyand, T.P.Speed, Comparison of discrimination methods for the classification of tumors using gene expression data, J.Am.Stat.Assoc.97 (2002) 77.
・R.Simon, M.Radmacher, K.Dobbin, L.M.McShane, Pitfalls in the use of DNA microarray data for diagnostic and prognostic classification, .Natl.Cancer Inst.95 (2003) 14.