[BioC] BioC] LIMMA: paired samples AND technical replicates

[Gordon Smyth]

Dear Charlotte,

Paired samples are normally handled as described in Section 8.3 
"Paired Samples" of the limma User's Guide. There shouldn't be any 
conflict between this and your technical replicates.

Best wishes
Gordon

>Date: Wed, 05 Sep 2007 17:54:42 +0200
>From: Charlotte Schjerling <araneus at mRNA.dk>
>Subject: [BioC] LIMMA: paired samples AND technical replicates
>To: bioconductor at stat.math.ethz.ch
>Message-ID: <20070905155442.ADC4E5C29 at ns1-int.rh.dk>
>Content-Type: text/plain;       charset=us-ascii;       format=flowed
>
>I have tried to search the BioC archives but have not found posts
>dealing with both paired samples and technical replicates in a setup
>like mine.
>
>I have 3 patients A, B, and C from which I have 4 samples: M.early,
>nM.early, M.late, and nM.late. Each sample has a technical
>replicate resulting in 24 arrays (Affymetrix).
>
>biolrep Patient M.early nM.early        M.late  nM.late
>1       A       1       0       0       0
>1       A       1       0       0       0
>2       A       0       1       0       0
>2       A       0       1       0       0
>3       A       0       0       1       0
>3       A       0       0       1       0
>4       A       0       0       0       1
>4       A       0       0       0       1
>5       B       1       0       0       0
>5       B       1       0       0       0
>6       B       0       1       0       0
>6       B       0       1       0       0
>7       B       0       0       1       0
>7       B       0       0       1       0
>8       B       0       0       0       1
>8       B       0       0       0       1
>9       C       1       0       0       0
>9       C       1       0       0       0
>10      C       0       1       0       0
>10      C       0       1       0       0
>11      C       0       0       1       0
>11      C       0       0       1       0
>12      C       0       0       0       1
>12      C       0       0       0       1
>
>I would like to find the following 4 contrasts:
>
>1) M.early - nM.early
>2) M.late - nM.late
>3) M.early - M.late
>4) nM.early - nM.late
>
>biolrep <- c(1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12)
>corfit <- duplicateCorrelation(eset, design, ndups=1, block=biolrep)
>
>If I ignore the pairing part I have constructed the following design matrix:
>
>  > design
>     M.early nM.early M.late nM.late
>1             1               0           0              0
>2             1               0           0              0
>3             0               1           0              0
>4             0               1           0              0
>5             0               0           1              0
>6             0               0           1              0
>7             0               0           0              1
>8             0               0           0              1
>9             1               0           0              0
>10            1               0           0              0
>11            0               1           0              0
>12            0               1           0              0
>13            0               0           1              0
>14            0               0           1              0
>15            0               0           0              1
>16            0               0           0              1
>17            1               0           0              0
>18            1               0           0              0
>19            0               1           0              0
>20            0               1           0              0
>21            0               0           1              0
>22            0               0           1              0
>23            0               0           0              1
>24            0               0           0              1
>attr(,"assign")
>[1] 1 1 1 1
>attr(,"contrasts")
>attr(,"contrasts")$f
>[1] "contr.treatment"
>
>
>fit <- lmFit(eset, design, ndups=1, block=biolrep, cor=corfit$consensus)
>contrasts <- makeContrasts(M.early-nM.early, M.late-nM.late,
>M.early-M.late, nM.early-nM.late, levels=design)
>fit2 <- contrasts.fit(fit, contrasts)
>fit2 <- eBayes(fit2)
>
>However, I would like to include the pairing by patient into the
>analyses and that I cannot figure out how to do.
>
>Hope someone can help,
>
>Charlotte