Dear Flavia,
You can look at
http://www.ncbi.nlm.nih.gov/pubmed/15929085
for some discussions of this topic and also at
http://www.ncbi.nlm.nih.gov/pubmed/23046981
for more general recommendations.
In general, eyeballing the data is a very helpful thing to do and you
should use the transform that emphasizes the effect you are interested
in and 'makes it look better' as a rule of thumb.
Vladimir
On Wed, Dec 19, 2012 at 4:00 PM, Cardini, Flavia
<[log in to unmask]> wrote:
> Hi,
> I am conducting a TF analysis and I would like to understand the difference
> between the several options I can choose to rescale my data. In particular,
> since I do not want to baseline-correct my data, I would like to understand
> when the "Log" function has to be used rather than the "Sqrt" and viceversa.
> I am sorry if the question sounds stupid but the manual does not provide any
> exhaustive explanation...
> is there any other manual or paper that could be useful to understand?
>
> flavia
|