[BioC] Fwd: statistical test for time course data

[Alex Gutteridge]

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