http://www.genomemedicine.com/content/5/10/92, this paper shows practical use of random forest Statistical software on microarray data. The codes can be retrieved from  GitHub https://github.com/obigriffith/biostar-tutorials/blob/master/MachineLearning/build_RF_model.R, The pros and cons are discussed in another article http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1363357/. Some of the highlighted parts from the above article is:::

Random forest has excellent performance in classification tasks, comparable to support vector machines. Although random forest is not widely used in the microarray literature (but see [18-23]), it has several characteristics that make it ideal for these data sets:

a) Can be used when there are many more variables than observations.

b) Can be used both for two-class and multi-class problems of more than two classes.

c) Has good predictive performance even when most predictive variables are noise, and therefore it does not require a pre-selection of genes (i.e., "shows strong robustness with respect to large feature sets", sensu [4]).

d) Does not overfit.

e) Can handle a mixture of categorical and continuous predictors.

f) Incorporates interactions among predictor variables.

g) The output is invariant to monotone transformations of the predictors.

h) There are high quality and free implementations: the original Fortran code from L. Breiman and A. Cutler, and an R package from A. Liaw and M. Wiener [24].

i) Returns measures of variable (gene) importance.

Hi,

you can also read the recent review paper http://www.lsta.upmc.fr/BIAU/bs.pdf with a complete list of references about RF, mathematical results and some practical consideration (algorithms, parameters tuning,variable selection, ...)

I have used them independently, but not together.  Are you asking about their intersection?