Entering edit mode
John Coulthard
▴
170
@john-coulthard-3077
Last seen 10.2 years ago
Hi
I'm having troubles with the removeBatchEffect function. It seems
that whatever I put in as the 'design' I get the same result out of
removeBatchEffect(); Perhaps I'm not using the right syntax for the
'design'?
Below is each line of the function run individually and the output
from the line using 'design' with various inputs; Always the same
output?!
My design is 4x2x2; 4 tissues(T) x 2 subgroups(S) x 2 days(D) and the
col order in e.qn is...
T1S1D1, T1S2D1, T1S1D2, T1S2D2, T2S1D1, T2S2D1, T2S1D2, T2S2D2,
T3S1D1, T3S2D1, T3S1D2, T3S2D2, T4S1D1, T4S2D1, T4S1D2, T4S2D2,
And it's the day(D) effect I'm trying to remove.
Can anyone point me to the right way to run this.
Many thanks
John
the command I started with...
e.qn_batch_removed<-removeBatchEffect(e.qn,c(1,1,2,2,1,1,2,2,1,1,2,2,1
,1,2,2),design=matrix(c(1,2,1,2,3,4,3,4,5,6,5,6,7,8,7,8),16,1))
then line by line...
> removeBatchEffect
function (x, batch, design = matrix(1, ncol(x), 1))
{
x <- as.matrix(x)
batch <- as.factor(batch)
contrasts(batch) <- contr.sum(levels(batch))
X <- model.matrix(~batch)[, -1, drop = FALSE]
X <- qr.resid(qr(design), X)
qrX <- qr(X)
t(qr.resid(qrX, t(x)))
}
<environment: namespace:limma="">
>
> x <- as.matrix(e.qn)
> batch <- as.factor(c(1,1,2,2,1,1,2,2,1,1,2,2,1,1,2,2))
> contrasts(batch) <- contr.sum(levels(batch))
> X <- model.matrix(~batch)[, -1, drop = FALSE]
> qr.resid(qr(matrix(c(1,2,1,2,3,4,3,4,5,6,5,6,7,8,7,8),16,1)), X)
batch1
1 1
2 1
3 -1
4 -1
5 1
6 1
7 -1
8 -1
9 1
10 1
11 -1
12 -1
13 1
14 1
15 -1
16 -1
> qr.resid(qr(matrix(c(1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2),16,1)), X)
batch1
1 1
2 1
3 -1
4 -1
5 1
6 1
7 -1
8 -1
9 1
10 1
11 -1
12 -1
13 1
14 1
15 -1
16 -1
> qr.resid(qr(matrix(1,16,1)), X)
batch1
1 1
2 1
3 -1
4 -1
5 1
6 1
7 -1
8 -1
9 1
10 1
11 -1
12 -1
13 1
14 1
15 -1
16 -1
>
[[alternative HTML version deleted]]
