prot2D (version 1.10.0)

variousT.Prot: Various Tests for finding differentially expressed proteins in 2-DE experiments

Description

These functions provides simple methods for finding differentially expressed proteins in 2-DE experiments.

Usage

ttest.Prot(data, plot = T, fdr.thr = 0.1, Fold2 = F, method.fdr = "BH", var.equal = F)
modT.Prot(data, plot = T, fdr.thr = 0.1, Fold2 = F, method.fdr = "BH", col=1)
samT.Prot(data, plot = T, fdr.thr = 0.1, Fold2 = F, method.fdr = "BH")
efronT.Prot(data, plot = T, fdr.thr = 0.1, Fold2 = F, method.fdr = "BH")
shrinkT.Prot(data, plot = T, fdr.thr = 0.1, Fold2 = F, method.fdr = "BH", var.equal = F)

Arguments

data

an ExpressionSet of volume data. Usually as returned by ES.prot.

fdr.thr

threshold value for the FDR (defaults to 0.1).

method.fdr

method for FDR estimator. 4 methods are implemented :"BH" for the Classical FDR from Benjamini & Hochberg, "Strimmer", "Storey" and "Pounds" (see references for details).

Fold2

logical. If true only spots with an absolute ratio of 2 are selected.

plot

logical. If true (default) draws a plot fdr values depending on the p-values with a line indicating the FDR threshold.

var.equal

logical. If false (default) assume unequal variances in each condition and uses a proper correction. Only for ttest.Prot and shrinkT.Prot.

col

An integer (default to 1). Only for modT.Prot. If there is more than one factor in pData, indicates the column to use for the analysis

Value

returns an ExpressionSet containing only the significant spots (see Examples).

Details

As described in Artigaud et al (2013) these are functions adapted from microarray analysis. 2-DE experiments analysis requires a variant of the t-statistic that is suitable for high-dimensional data and large-scale multiple testing. For this purpose, in the last few years, various test procedures have been suggested. These functions provides: - the classical Student's t-test (adapted from studentt.stat). - two tests especially modified for micro-array analysis : Efron's t-test (adapted from efront.stat, Efron et al, 2001) and the modified t-test used in Significance Analysis for Microarray (adapted from samr, Tusher et al, 2001) - two methods that take advantage of hierarchical Bayes methods for estimation of the variance across genes: the moderate t-test from Smyth (using limma; see Smyth, 2004) and the "Shrinkage t" statistic test from Opgen-Rhein & Strimmer (adapted from shrinkcat.stat; see Opgen-Rhein & Strimmer, 2007). As statistical tests allowing the identification of differentially expressed proteins must take into account a correction for multiple tests in order to avoid false conclusions. These functions also provides different methods to estimate the False Discovery Rate : - the classical FDR estimator of Benjamini & Hochberg (using p.adjust; see Benjamini & Hochberg, 1995) - the Fdr estimator of Strimmer (based on local fdr calculation) (using fdrtool; see Strimmer 2008) - the "robust FDR" estimator of Pounds & Cheng (implemented in robust.fdr for the prot2D package; see Pounds & Cheng, 2006) - Fdr method of Storey and Tibshirani (2003), also known as "q-values" (using qvalue.

References

• Artigaud, S., Gauthier, O. & Pichereau, V. (2013) "Identifying differentially expressed proteins in two-dimensional electrophoresis experiments: inputs from transcriptomics statistical tools." Bioinformatics, vol.29 (21): 2729-2734.
• Benjamini, Y. & Hochberg, Y. (1995) "Controlling the false discovery rate: a practical and powerful approach to multiple testing" Journal of the Royal Statistical Society. Series B. Methodological.: 289-300.
• Efron, B., Tibshirani, R., Storey, J.D., & Tusher, V. (2001) "Empirical Bayes Analysis of a Microarray Experiment" Journal of the American Statistical Association, vol. 96 (456): 1151-1160.
• Tusher, V.G., Tibshirani, R., & Chu, G. (2001) "Significance analysis of microarrays applied to the ionizing radiation response"" Proceedings of the National Academy of Sciences of the United States of America, vol. 98 (9): 5116-5121.
• Smyth, G.K. (2004) "Linear models and empirical bayes methods for assessing differential expression in microarray experiments." Statistical Applications in Genetics and Molecular Biology, vol. 3: Article 3.
• Pounds, S. & Cheng, C. (2006) "Robust estimation of the false discovery rate" Bioinformatics, vol. 22 (16): 1979-1987.
• Strimmer, K. (2008) "A unified approach to false discovery rate estimation." BMC Bioinformatics, vol. 9: 303.
• Opgen-Rhein, R. & Strimmer, K. (2007) "Accurate Ranking of Differentially Expressed Genes by a Distribution-Free Shrinkage Approach" Statistical Applications in Genetics and Molecular Biology, vol. 6 (1).

See Also

Norm.qt,ES.prot,fdrtool,limma, samr,studentt.stat,shrinkt.stat, efront.stat,qvalue

Examples

data(pecten)
data(pecten.fac)

pecten.norm <- Norm.qt(pecten, n1=6, n2=6, plot=TRUE) #Quantiles normalization of the data
ES.p <- ES.prot(pecten.norm, n1=6, n2=6, f=pecten.fac)
ES.diff <- modT.Prot(ES.p, fdr.thr=0.1, plot=TRUE)
featureNames(ES.diff) # Names of the spots selected for a moderated t-test with a fdr of 0.1
fData(ES.diff) # Displaying fold change (as log2(ratio)) for selected spots
exprs(ES.diff) # Volume normalized data for all the selected spots
## Not run: heatplot(ES.diff) #Great heatmap of the selected spots (require made4 Bioconductor package )