[BioC] timecourse: modt1 not found

[Steve Ladunga]

Hi,

I got the error message:
> out1 <- mb.long(foldM,times = 11, time.grp=times, 
reps = size1, rep.grp =
repl, out.t=TRUE,method="1D")

Error in mb.1D(object, times, reps, prior.df,
prior.COV, prior.eta, time.grp,  : 
  object 'modt1' not found


We have a one-condition 11 time point experiment 
with 2-color custom microarrays in 6 replicates.

Background-correction and normalization 
was performed in limma, dye-swaps were
taken into account by multiplying their columns by -1.

The fold change values were extracted into a 
matrix object foldM

> dim(foldM)
[1] 17328    66
> times
 [1]  9  9  9  9  9  9 16 16 16 16 16 16 23 23 23 23
 23 23 30 30 30 30 30 30 37
[26] 37 37 37 37 37 44 44 44 44 44 44 51 51 51 51 51
51 58 58 58 58 58 58 65 65
[51] 65 65 65 65 72 72 72 72 72 72 79 79 79 79 79 79
> repl
 [1] "A" "B" "C" "D" "E" "F" "A" "B" "C" "D" "E" "F" 
"A" "B" "C" "D" "E" "F" "A"
[20] "B" "C" "D" "E" "F" "A" "B" "C" "D" "E" "F" "A"
 "B" "C" "D" "E" "F" "A" "B"
[39] "C" "D" "E" "F" "A" "B" "C" "D" "E" "F" "A" "B" 
"C" "D" "E" "F" "A" "B" "C"
[58] "D" "E" "F" "A" "B" "C" "D" "E" "F"
> size1<- rep(6,17328)

I run timecourse on a 64-bit CentOS LINUX.

I guess the problem can be in this part of mB.1D.R:

    122    tol <- .Machine$double.eps
    123    if(is.finite(nu) & nu > tol) 
    124    {
    125      modS1 <- sapply(1:G, function(x)
((n[x]-1)*matrix(S1[,x],ncol=k-1)+nu*Lambda1)/(n[x]-1+nu))
    126      if(is.null(eta) & (T2.only==FALSE))
    127      {
    128        sqrt.modS1 <- apply(modS1, 2, function(x)
    129       
svd(matrix(x,ncol=k-1))$u%*%diag(sqrt

(svd(matrix(x,ncol=k-1))$d))%*%t(svd(matrix(x,ncol=k-1))$v))

    130        modt1 <- sapply(1:G, function(x)
n[x]^(1/2)*solve(matrix(sqrt.modS1[,x],
ncol=k-1))%*%T1%*%xbar[,x])
    131        HotellingT2 <- apply(modt1, 2, 
function(x) t(x)%*%x)
    132      }
    133      if(T2.only)
    134          HotellingT2 <- sapply(1:G, function(x) 
    135                
n[x]*t(T1%*%xbar[,x])%*%solve(matrix(modS1[,x],ncol=k-1))

%*%T1%*%xbar[,x])
    136 
    137       if(!is.null(eta) & (T2.only==FALSE))
    138          HotellingT2 <- sapply(1:G, function(x)
    139                
n[x]*t(T1%*%xbar[,x])%*%solve(matrix(modS1[,x],
ncol=k-1))%*%T1%*%xbar[,x])
    140 
    141    }
 

Any help would be appreciated!

Thank you so much!

Steve
Steve Ladunga
Department of Statistics
University of Nebraska-Lincoln