The study of Coffey et al. (2009) compares the effect of diverse contractile activity in human skeletal muscle. They measured phosphorylation of several proteins and the results (phospho/total) are log-transformed. I don't understand the interest of these method.
Hi Frederick, one reason to log-transform the data is that upregulated or downregulated data (compared to a control sample) will not be normally distributed, which can be problematic if you want to use a parametric statistical test (t-test or ANOVA for example). The reason is that upregulated data can have values from 1 up to infinity while downregulated data will have values from 1 to 0 on a linear scale (4 times upregulation has a value of 4 while 4 time downregulation has a value of 0.25). By log transforming the data, upregulated and downregulated expression data can then have a similar scale of changes.
Hope that make sense
http://errantscience.com/blog/2014/09/17/an-infographic-guide-to-data/
(but otherwise it's as Claude said:)
- Badges
- Science method