mdPat {mdgsa} | R Documentation |
Classifies significant patterns form a Multi-Variate Gene Set Analysis.
mdPat(gsaout, cutoff = 0.05, pvalue = "padj")
gsaout |
data.frame; output from mdGsa. |
cutoff |
p-value cutoff for considering significant a Gene Set. |
pvalue |
p-value column to be used. Default is named "padj" as in mdGsa output. |
Sign of the three 'lor' and p-values are used to classify functional blocks. The classification is done in the two dimensional space previously analyzed by mdGsa.
All possible functional block classifications in the bi-dimensional gene set analysis are:
q1i: block displaced toward quadrant 1 (0 < X & 0 < Y) with interaction.
q2i: block displaced toward quadrant 2 (0 > X & 0 < Y) with interaction.
q3i: block displaced toward quadrant 3 (0 > X & 0 > Y) with interaction.
q4i: block displaced toward quadrant 4 (0 < X & 0 > Y) with interaction.
q1f: block displaced toward quadrant 1, no interaction.
q2f: block displaced toward quadrant 2, no interaction.
q3f: block displaced toward quadrant 3, no interaction.
q4f: block displaced toward quadrant 4, no interaction.
xh: block shifted to positive X values.
xl: block shifted to negative X values.
yh: block shifted to positive Y values.
yl: block shifted to negative Y values.
b13: bimodal block. Half of the genes displaced towards quadrant 1 and the other half towards quadrant 3.
b24: bimodal block. Half of the genes displaced towards quadrant 2 and the other half towards quadrant 4.
NS: non significant block.
A character vector indicating the pattern associated to each Gene Set.
David Montaner dmontaner@cipf.es
Montaner et al. (2010) "Multidimensional Gene Set Analysis of Genomic Data." PLoS ONE.
N <- c (10, 20, 30, 40) lor.X <- c (1.45, -0.32, 1.89, -1.66) lor.Y <- c (2.36, -1.86, 0.43, -2.01) lor.I <- c (0.89, -0.12, 0.24, 3.55) pval.X <- c (0.001, 0.002, 0.003, 0.06) pval.Y <- c (0.002, 0.003, 0.06, 0.07) pval.I <- c (0.003, 0.02, 0.05, 0.08) padj.X <- p.adjust (pval.X, "BY") padj.Y <- p.adjust (pval.Y, "BY") padj.I <- p.adjust (pval.I, "BY") mdGsa.res <- as.data.frame (cbind (N, lor.X, lor.Y, lor.I, pval.X, pval.Y, pval.I, padj.X, padj.Y, padj.I)) mdGsa.res mdGsa.res[,"pat"] <- mdPat (mdGsa.res) mdGsa.res