[BioC] Problem running eBayes/lmFit

[smohapat at vbi.vt.edu]

Hello Elliott:

I am trying to understand the problem here. From the design:

> A9802 A9811 A9813 A9842 SAMPLE1 SAMPLE2 SAMPLE3 SAMPLE4
> 1     0     0     0     0       1       0       0       0
> 2     0     0     0     0       0       1       0       0
> 3     0     0     0     0       0       0       1       0
> 4     0     0     0     0       0       0       0       1
> 5     1     0     0     0       0       0       0       0
> 6     0     1     0     0       0       0       0       0
> 7     0     0     1     0       0       0       0       0
> 8     0     0     0     1       0       0       0       0

I understand that there is one sample of A9802(#5) and another of SAMPLE1
(#1).

In the contrasts, these two groups are compared (OneVOne):
> makeContrasts(OneVOne="A9802-SAMPLE1",OneVTwo="A9802-A9811",TwoVOne="SAM
> PLE1-SAMPLE2",levels=design)
>

I guess that because of the number of samples in each group being one, it
is not possible to calculate variance, and hence the error message.

This is how I understood Gordon's earlier post
(https://stat.ethz.ch/pipermail/bioconductor/2005-May/009056.html):

------------
The "no residual degrees of freedom" message occurs because you have
filtered out so many spots
(by setting the weight to 0) that you have no more than one spot left for
any of the probes.
Hence there is no replication left in your experiment.  No estimate of
variability can be made and
no statistical analysis can be done.
-------------

If anyone knows more clearly, please elaborate.

Saroj

> The matrix seems to be doing what I want
>
>
>> cont.matrix
> Contrasts
> Levels    OneVOne OneVTwo TwoVOne
> A9802         1       1       0
> A9811         0      -1       0
> A9813         0       0       0
> A9842         0       0       0
> SAMPLE1      -1       0       1
> SAMPLE2       0       0      -1
> SAMPLE3       0       0       0
> SAMPLE4       0       0       0
>
>
>> fit2  <- contrasts.fit(fit, cont.matrix) fit2  <- eBayes(fit2)
> Error in ebayes(fit = fit, proportion = proportion, stdev.coef.lim =
> stdev.coef.lim) : No residual degrees of freedom in linear model fits
>
>
> I've found a post that says this error message occurs because all data
> is weighted out. I've checked the data after it is loaded, after
> backgroundCorrect and it does not appear to be. Beyond that I doesn't look
> like the normalizeBetweenArrays of RTotalbg$R RTRN has any weights. So I
> must not be setting up the design matrix correctly?
>
> Any and all clues as to where I'm going wrong greatly appreciated.
>
>
>
> Elliott Harrison
>
>
>
>
>
> This message has been scanned for viruses by BlackSpider...{{dropped:3}}
>
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
>
>