Cuffdiff no without replicates

classic Classic list List threaded Threaded
3 messages Options
i b
Reply | Threaded
Open this post in threaded view
|

Cuffdiff no without replicates

i b
Dear all,
how reliable is running Cuffdiff without replicates? e.g.one samples
agains another one?

Is it statistically makign any difference when using replicates?

Thanks,
ib
___________________________________________________________
The Galaxy User list should be used for the discussion of
Galaxy analysis and other features on the public server
at usegalaxy.org.  Please keep all replies on the list by
using "reply all" in your mail client.  For discussion of
local Galaxy instances and the Galaxy source code, please
use the Galaxy Development list:

  http://lists.bx.psu.edu/listinfo/galaxy-dev

To manage your subscriptions to this and other Galaxy lists,
please use the interface at:

  http://lists.bx.psu.edu/
Reply | Threaded
Open this post in threaded view
|

Re: Cuffdiff no without replicates

fubar
On Wed, Oct 3, 2012 at 9:11 PM, i b <[hidden email]> wrote:
> Dear all,
> how reliable is running Cuffdiff without replicates? e.g.one samples
> agains another one?
>
> Is it statistically makign any difference when using replicates?

Seqanswers might be a better place to ask this very interesting
technical question that goes way beyond Galaxy...

My 2c: Statistically speaking, sequencing and biology are both noisy.
Replicates provide information about non-experimental (technical and
biological) variation. That variation is usually not the variation you
are looking for, but if you want to remove it, you have to model it
and that requires information from replicates (or really good
guesswork). In some situations (eg extreme experimental conditions),
I'm sure you'll find biologically meaningful signal without them but
in my experience, they can really help to decrease non-experimental
noise, particularly where the experimental condition induces only
subtle changes in transcript abundance.

You could always analyse a data set with replicates and compare the
results with and without those replicates yourself to see what happens
- it would be a nice paper I'm sure.

>
> Thanks,
> ib
___________________________________________________________
The Galaxy User list should be used for the discussion of
Galaxy analysis and other features on the public server
at usegalaxy.org.  Please keep all replies on the list by
using "reply all" in your mail client.  For discussion of
local Galaxy instances and the Galaxy source code, please
use the Galaxy Development list:

  http://lists.bx.psu.edu/listinfo/galaxy-dev

To manage your subscriptions to this and other Galaxy lists,
please use the interface at:

  http://lists.bx.psu.edu/
Reply | Threaded
Open this post in threaded view
|

Re: Cuffdiff no without replicates

Sean Davis
On Wed, Oct 3, 2012 at 7:35 AM, Ross <[hidden email]> wrote:

> On Wed, Oct 3, 2012 at 9:11 PM, i b <[hidden email]> wrote:
>> Dear all,
>> how reliable is running Cuffdiff without replicates? e.g.one samples
>> agains another one?
>>
>> Is it statistically makign any difference when using replicates?
>
> Seqanswers might be a better place to ask this very interesting
> technical question that goes way beyond Galaxy...
>
> My 2c: Statistically speaking, sequencing and biology are both noisy.
> Replicates provide information about non-experimental (technical and
> biological) variation. That variation is usually not the variation you
> are looking for, but if you want to remove it, you have to model it
> and that requires information from replicates (or really good
> guesswork). In some situations (eg extreme experimental conditions),
> I'm sure you'll find biologically meaningful signal without them but
> in my experience, they can really help to decrease non-experimental
> noise, particularly where the experimental condition induces only
> subtle changes in transcript abundance.
>
> You could always analyse a data set with replicates and compare the
> results with and without those replicates yourself to see what happens
> - it would be a nice paper I'm sure.

A bit off-topic, but you might take a look here:

http://www.ncbi.nlm.nih.gov/pubmed/21747377

In short, one needs replication in biology, regardless of the
technology used.  In particular, one would never suggest running a
microarray experiment without replicates; one should follow
approximately the same rules for sequencing (and sequence data
analysis).

Sean

>>
>> Thanks,
>> ib
> ___________________________________________________________
> The Galaxy User list should be used for the discussion of
> Galaxy analysis and other features on the public server
> at usegalaxy.org.  Please keep all replies on the list by
> using "reply all" in your mail client.  For discussion of
> local Galaxy instances and the Galaxy source code, please
> use the Galaxy Development list:
>
>   http://lists.bx.psu.edu/listinfo/galaxy-dev
>
> To manage your subscriptions to this and other Galaxy lists,
> please use the interface at:
>
>   http://lists.bx.psu.edu/
___________________________________________________________
The Galaxy User list should be used for the discussion of
Galaxy analysis and other features on the public server
at usegalaxy.org.  Please keep all replies on the list by
using "reply all" in your mail client.  For discussion of
local Galaxy instances and the Galaxy source code, please
use the Galaxy Development list:

  http://lists.bx.psu.edu/listinfo/galaxy-dev

To manage your subscriptions to this and other Galaxy lists,
please use the interface at:

  http://lists.bx.psu.edu/