UsingR

While AltAnalyze is a largely a stand-alone program, some statistical analyses can be included that depend on external applications. These require prior installation of these tools using operating system installers and properly interfacing them with AltAnalyze. For Windows operating system, R is already included with the binary version of AltAnalyze. This allows the software to immediately use HOPACH clustering and to perform Monocle.

Interactivity with R

Although installation of R is not required for any of the AltAnalyze analyses, for users who wish to use more advanced statistics (e.g., advanced clustering), installation of R may be needed. AltAnalyze connects to R using the python library pypeR (PYthon-piPE-R), included with AltAnalyze. Using this library, AltAnalyze is able to connect to R, access libraries and run them remotely. Currently, clustering using the hopach algorithm is the only R library supported. To run, you will only need to have R installed and recognized from a terminal window. Most R installers should include this functionality by default, without any user-required interaction. AltAnalyze will install hopach in a local directory when first run. R interactivity is supplied using the R_interface.py module. Additional functionality has been built into this module (gcrma, multtest, combat, limma), but is not currently connected to any main analysis options within AltAnalyze. Advanced users wishing to exploit this code are welcome to contribute to AltAnalyze development. Please note, that some limma and combat functionality are already supported using existing python libraries in AltAnalyze.