Personally I find Python to be a great general purpose cross platform language (see my Python Pages). However, R has its strengths too, particularly in statistics and graphics.

Calling R from Python

R's rich libraries for statistics and graph creation can be called from within a Python program using RPy (R from Python), written by Walter Moreira and Gregory Warnes. It works on both Linux and Windows.

I have used this to:

• Calculate linear regressions
• As one of several ways to calculate rank correlations.
• Draw heatmaps from Python
• Draw Ramachandran Plots from Python, using filled contour plots with an overlayed scatter plot described here in just R.
• Draing protein contact maps, another contour plot example.

See my python pages for further examples...

Calling Python from R

This should be possible using RSPython, an R/SPlus - Python Interface. I have not tried this yet, mainly because of their weak Windows support. On linux at least, RSPython should allow you to call R from Python, and Python from R - which makes it rather more powerful than RPy.

The R Project

 

Python

 

RPy (R from Python)