SMART

1. How can I install this software?

The source code of SMART is stored on PyPI - the Python Package Index which is a repository of software for the Python programming language. To use SMART package from this index either "pip install SMART-BS-Seq" (get pip) or download, unpack and "python setup.py install" it.
See Install page for detail.

2. How much space and memory will it occupy?

Due to the large-scale of BS-Seq data and Cytosine location files (for instance, these files is about 4G) and the principle of giving user intermediate results as much as possible, this software needs tree times of the space which is occupied by your methylation data. So for instance, 50 BS-Seq methylomes (formated in wig.gz) is about 10G, the result folder would be about 14G. Considering the space of Cytosine location files (for instance, these files is about 4G), this software need about 28G. As SMART uses line processing mode, thus it does not consume much memory. Take the same example of 50 BS-Seq methylomes, this software only use 25M memory.

3. How long will it take to run an analysis?

For 50 BS-Seq methylomes (formated in wig.gz, about 10G) on human, it will take less than 50 hour and no more than 30M memory to complete the analysis on a 3.2G Hz computer, using the default parameter. This project report can be download here for reference. Current, we are trying to shorten this process.

4. What operating systems are needed to use this software?

Currently, both MS windows 64 bit and Linux are suported by SMART. We have test this software in Window8 (64 bit)and two Linux systems: Ubuntu 12.04 and Red Hat Enterprise Linux Server release 6.2.

5. What python packages are needed to use this software?

In addition Python 2.7, the Scipy is needed for this software. Under windows, Scipy can be obtained from http://www.lfd.uci.edu/~gohlke/pythonlibs/#scipy. For Linux, Scipy can be installed by "pip install Scipy". Sometimes, numpy may also needed before installing Scipy. The numpy can be installed by "pip install numpy"

 

For any help:	you are welcome to write to Hongbo Liu (hongbo919@gmail.com).