am.trans: Transform replicated arrays into (A,M) format

Description

Transforms expression intensity measurements for replicated arrays of a single experimental condition into (A,M) format: A = (xi+xj)/2, M = (xi-xj) where x1, x2,..., xn are individual chips. This function is used in the estimation of within-bin variances in the LPE function, and not typically by the user.

Usage

am.trans(y)

Arguments

y is an ngene by n matrix of expression intensity measurements for replicated arrays under a single experimental condition.

Value

Returns matrix with 2 columns cbind(A.M) and rows comprising all permutations of individual chip columns of the input matrix y. Note that for each pair of chips M is calculated twice, once for xi-xj and once for xj-xi. The resulting matrix thus has twice the number of rows as the input matrix y.

References

J.K. Lee and M.O.Connell(2003). An S-Plus library for the analysis of differential expression. In The Analysis of Gene Expression Data: Methods and Software. Edited by G. Parmigiani, ES Garrett, RA Irizarry ad SL Zegar. Springer, NewYork.

Jain et. al. (2003) Local pooled error test for identifying differentially expressed genes with a small number of replicated microarrays, Bioinformatics, 1945-1951.

Jain et. al. (2005) Rank-invariant resampling based estimation of false discovery rate for analysis of small sample microarray data, BMC Bioinformatics, Vol 6, 187.

Examples

Run this code

  library(LPE)
  # Loading the LPE library
 
  # Test data with duplicates
  temp1 <- matrix(c(1,20,1.5,23),nrow=2)
  am.trans(temp1)
  # It gives a matrix of (4*2) as only two permutaions
  # are possible for each row with duplicates (X1-X2, and X2-X1)
  
  
  # Another test data with three replicates
  temp2 <- matrix(c(1,20,1.5,23,0.8,19),nrow=2)
  am.trans(temp2)
  # Now it returns matrix of (12*2) as there are
  # 6 possible permutations for each row with triplicates
  # (X1-X2, X1-X3, X2-X3, X2-X1, X3-X1 and X3-X2)