Team:Illinois-Tools/Notebook/Week2


 * back to notebook

<>

5/31-6/6
Donny

This week, I learned the basics of the python programming language. I spend most of my time working with the Kegg API to access data via the SOAP protocol. I was able to utilize this new information to create a module that converts the common name to a usable kegg_id and also to convert the kegg_id back to a name. One concern I have is that even these simple functions are somewhat slow. Perhaps we will find ways to speed these processes up in the weeks to come. I looked into the possibility of Google App Engine. The one shortcoming (although it is pretty cool) was that there were some restrictions as to which libraries can be used. I foresee this as a problem and am looking into other options.

Riyad

We got Cobra working so I spent the majority of the week getting used to how Python and Cobra worked.

Palak

I began working with the various functions in COBRA. I spent a lot of time understanding the output of COBRA and determining what aspects are necessary in creating an optimal pathway. I also looked at the inputs required for various COBRA functions and looked into ways that we can pass information from the KEGG database into COBRA. I figured out how to use the KEGG API. Additionally, I spend some time looking at the COBRA source code to see what we could modify to fit the needs of our program.

Nate

Since last week, I had some difficulties with installing python modules, specifically the ones that would allow us to graphically represent our program's outputs. Tried to get a program called Easy_Install to work, which would make installing modules easy, but it was difficult to figure out how to get Easy_Install to install... Fortunately, Donny was able to find an easy way to install the modules via MS-DOS prompt on windows. Now I am able to install any module...almost. I ended up installing these modules:

-Networkx

-Pydot

-Matplotlib

As well as these additional programs:

-Graphviz

-Cytoscape

Up to this point we have been using Python 2.6, but unfortunately I was unable to get all of the above modules to install correctly for Python 2.6. I downgraded to Python 2.5, and was able to get all of the modules to install and work properly. I was finally able to generate a graph via Python in conjunction with Graphviz!! I also experimented more with Cytoscape, another graphing program that we could use. However, Cytoscape does not appear to be compatible with Python at the moment. We probably will not be using Cytoscape.

Kanishka

Spent most of the time downloading different tools in order to get python and additional features to work correctly and looked for a good tool to graphically represent the network.