[BioC] Fwd: statistical test for time course data
Alex Gutteridge
alexg at ruggedtextile.com
Mon Feb 18 12:11:51 CET 2013
On 18.02.2013 07:17, chris Jhon wrote:
> Hi All,
>
> I appreciate any help.
>
>
> Hi ;
>
> Thank you Richard for help.
> I have the data like this table
>
> Time number
> 0hr #
> 6hr #
> 24hr #
>
> i tried to follow the example as in userguide and as Richard
> suggested
> me,but have the following questions:
> in user guide
> ***************
>> lev <- c("wt.0hr","wt.6hr","wt.24hr","mu.0hr","mu.6hr","mu.24hr")
>> f <- factor(targets$Target, levels=lev)
>> design <- model.matrix(~0+f)
>> colnames(design) <- lev
>> fit <- lmFit(eset, design)
> ***************
>
> Q1) what about est, in this stage i would like to test the
> statistical
> significance between numbers showed in second column which represents
> the
> number of expressed genes,SHALL I REPLACE ESET WITH MYDATA$number??
>
> when i tried so i got the following error --- Error in
> rowMeans(y$exprs,
> na.rm = TRUE) : 'x' must be numeric
>
> Q2) Can anyone explain for methe meaning of (~0+f) in
> design <- model.matrix(~0+f)
>
> Q3) how to design different matrices for different conditions,can any
> one
> send me a tutorial for this.
>
> Thank you very much in advance.
For Q2&3 I don't have any better suggestion that re-reading the Limma
users guide or some general introductory texts for statistical modelling
with R.
For Q1, if you really want to test whether the *number of expressed
genes* is different between samples (time points) (i.e. not differential
expression) and you have no replicates (?) then I don't see what you can
do apart from a binomial proportions test.
i.e. if the total number of genes in your studied system is 30,000 and
the number of genes 'expressed' at each of your three time points was
3100, 3000 and 4000 and you could try:
prop.test(c(3000,4000),c(30000,30000))
Which would show you that, yes indeed, 4000/30000 is a significantly
higher proportion than 3000/30000, but I'm really not sure if that is
what you actually want to do! It's not a common use case and from the
rest of your question I suspect there is some confusion with terminology
going on (no offense!). Personally, I would say this is one of those
times where you would be best served by sitting down with a friendly
local expert.
--
Alex Gutteridge
More information about the Bioconductor
mailing list