R help - Sep 2011 - Questions on 'lme' function, urgent!

Hi Dear all,

I have some gene expression data samples from different tissue types
-----------------------------------------------
- 120 samples from blood (B)
- 20 samples from Liver (L)
- 15 samples from Kidney (K)
- 6 samples from heart (H)
-----------------------------------------------
All the samples are from different individuals, so there are in total 161
individuals from which the DNA was collected.

For each sample DNA, the expression level of 200 genes were obtained, so the
whole expression data is a 161 x 200 matrix. 

The purpose of my project is to interrogate, across the 200 genes, how many
of them have an expression level consistent across all the tissue types
examined, how many of them have an expression level different between any
two of the tissue types, how many of them have an expression level unique to
only one tissue (i.e., gene20 is expressed at a very low level (0.1) in
Liver, but expressed at the same level (~0.7) across all the other tissue
types). 
So my hypotheses are:
H0: Mean(B)=Mean(L)=Mean(K)=Mean(H)
H1: Mean(B)?Mean(L)=Mean(K)=Mean(H) or Mean(L)?Mean(B)=Mean(K)=Mean(H) or    
      Mean(K)?Mean(B)=Mean(L)=Mean(H) or Mean(H)?Mean(B)=Mean(K)=Mean(L)
H2: Mean(B)?Mean(L)?Mean(K)?Mean(H)

In my analysis, the gene expression level is the dependent value (Y), the
tissue-type is the fixed effects, the inter-individual variation and the
batch effects are the random effects. So I was suggested to use the
'lme'
function to do the analysis. 
We want to compare the Likelihood ratio between models based on different
hypothesis. i.e., if I want to see, for one gene, if the expression levels
are all different between any two tissue types, that is to compare H2 to H1,
if the p value corresponding to the
D_H2H1=likelihoodRatio(H2)-likelihoodRatio(H1) in a chi-square distribution
is less than 0.05, then we will say H2 is accepted. 

However, my problem is, I don't know how to specify these parameterizations
in 'lme' function based H0, H1 and H2, respectively. Can anyone help me
with
that?

Thank you very much,

Karena



--
View this message in context:
http://r.789695.n4.nabble.com/Questions-on-lme-function-urgent-tp3816741p3816741.html
Sent from the R help mailing list archive at Nabble.com.

You should post this to the r-sig-mixed-models  list, not here.

-- Bert

On Thu, Sep 15, 2011 at 1:42 PM, karena <dr.jzhou@gmail.com> wrote:
> Hi Dear all,
>
> I have some gene expression data samples from different tissue types
> -----------------------------------------------
> - 120 samples from blood (B)
> - 20 samples from Liver (L)
> - 15 samples from Kidney (K)
> - 6 samples from heart (H)
> -----------------------------------------------
> All the samples are from different individuals, so there are in total 161
> individuals from which the DNA was collected.
>
> For each sample DNA, the expression level of 200 genes were obtained, so
> the
> whole expression data is a 161 x 200 matrix.
>
> The purpose of my project is to interrogate, across the 200 genes, how many
> of them have an expression level consistent across all the tissue types
> examined, how many of them have an expression level different between any
> two of the tissue types, how many of them have an expression level unique
> to
> only one tissue (i.e., gene20 is expressed at a very low level (0.1) in
> Liver, but expressed at the same level (~0.7) across all the other tissue
> types).
> So my hypotheses are:
> H0: Mean(B)=Mean(L)=Mean(K)=Mean(H)
> H1: Mean(B)≠Mean(L)=Mean(K)=Mean(H) or Mean(L)≠Mean(B)=Mean(K)=Mean(H) or
>      Mean(K)≠Mean(B)=Mean(L)=Mean(H) or Mean(H)≠Mean(B)=Mean(K)=Mean(L)
> H2: Mean(B)≠Mean(L)≠Mean(K)≠Mean(H)
>
> In my analysis, the gene expression level is the dependent value (Y), the
> tissue-type is the fixed effects, the inter-individual variation and the
> batch effects are the random effects. So I was suggested to use the
'lme'
> function to do the analysis.
> We want to compare the Likelihood ratio between models based on different
> hypothesis. i.e., if I want to see, for one gene, if the expression levels
> are all different between any two tissue types, that is to compare H2 to
> H1,
> if the p value corresponding to the
> D_H2H1=likelihoodRatio(H2)-likelihoodRatio(H1) in a chi-square distribution
> is less than 0.05, then we will say H2 is accepted.
>
> However, my problem is, I don't know how to specify these
parameterizations
> in 'lme' function based H0, H1 and H2, respectively. Can anyone
help me
> with
> that?
>
> Thank you very much,
>
> Karena
>
>
>
> --
> View this message in context:
>
http://r.789695.n4.nabble.com/Questions-on-lme-function-urgent-tp3816741p3816741.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


-- 
"Men by nature long to get on to the ultimate truths, and will often be
impatient with elementary studies or fight shy of them. If it were possible
to reach the ultimate truths without the elementary studies usually prefixed
to them, these would not be preparatory studies but superfluous
diversions."

-- Maimonides (1135-1204)

Bert Gunter
Genentech Nonclinical Biostatistics

	[[alternative HTML version deleted]]