[BioC] RNAseq expression analysis using DESeq: technical replicates, paired samples

[Michael Muratet]

On Nov 7, 2011, at 4:11 PM, Simon Anders wrote:

> Dear Michael
>
> On 2011-11-07 22:56, Michael Muratet wrote:
>> I would like to verify that 18 months later, adding counts for  
>> technical
>> replicates is still the best approach for combining technical  
>> replicates.
>>
>> We are constrained to single biological replicates, but we're  
>> interested
>> in using the technical replicates if we can. I recall that limma had
>> ways to set up the model matrix.
>
> Your question sounds as if you consider the issue as a technical  
> limitation of the available methods. Unfortunately, it is not, and I  
> am afraid, 18 months are not enough time for fundamental principles  
> of statistics to change.
Hi Simon

>
> If you want to know whether a difference between control and  
> treatment may be attributed to the treatment, you need to verify  
> that this difference is large compared to what you expect from  
> random biological variation, and without biological replicates you  
> cannot know the extent of biological variation.

No, I really was just checking for updates in how to arrange the  
input. I understand the limitations of no biological replicates.  
However, as noted in section 6 of the vignette, '...such experiments  
are often undertaken...' and we hope to confirm what we find by other  
means.

I thought I had seen documentation somewhere on how to incorporate  
technical replicates into deseq, but I could not put my hands back on  
it, so I thought I'd ask. I have actually processed the original data  
once and got what results we could get. I don't expect the tech  
replicates to change the answer.

Cheers

Mike

>>
> Of course, you can perform the analysis on the level of technical  
> replicates, as you can with microarrays in limma. Then, you will get  
> lots of results, but any call of significant differences may only be  
> interpreted to mean that your two biological samples differ in this  
> gene. You may not infer that the difference is more likely to be the  
> result of your treatment than just due to chance differences between  
> the samples.
>
> Best regards
>  Simon

Michael Muratet, Ph.D.
Senior Scientist
HudsonAlpha Institute for Biotechnology
mmuratet at hudsonalpha.org
(256) 327-0473 (p)
(256) 327-0966 (f)

Room 4005
601 Genome Way
Huntsville, Alabama 35806