From 2010.igem.org
- The notebook is organized into weeks. Click on the Saturday of each week to view that week's notebook entry
Weekly Lab Notebook
May 30 – June 5
Colin - This week, I have been more than a bit under the weather, and thus, have not been able to get as much done as I hoped. However, this week we determined the name for our project. The name is BioMORTAR, which stands for Metabolic Optimization Research Tool And Registry. After deciding between several names, this one was picked because it relates to the focus of our project: the automated design of a plasmid using BioBricks to construct it. Another break through with the project is the fact that we might have the chance, in the fall, to present our project at the Chemical and Biological Defense Conference. Since the application to the CBD S&T, requires an abstract to be submitted, I have been working, this week, on getting the abstract to be good enough to present to the Department of Defense.
Finally, we have decided this week that we will be needing a server dedicated to our program, just in case of high traffic and because we do not like having to bother Dr. Bhalerao by putting our stuff on his server.
Jeremy - I spent this week upgrading my integration of RPAIR. The database had to be manually downloaded, manually converted, and manually integrated into the main mysql database. I wrote some functions to automate these processes. While not strictly necessary, I was still feeling my way around django.
Bobak - Team tutorials in python and django
Connecting IMPtools pathway results to strain designer
Updating reaction database from KEGG, migrating to mysql
Working with Jeremy on rpair for reaction coordination
Anna - Continued to learn about Django this week. Django is able to take each application and gives a
view file. This view file is where all the functions are carried out and each function references a unique
page request within the website. We also learned more more about how IMPTools functions. To find
a reaction pathway, IMPTools uses Dijkstra’s algorithm by making compounds nodes and edges the
reactions between them. It takes the input and out and sets it as a start and end and looks at both
points in a bidirectional path and eventually the paths overlap which gives you the best first search.
Within each step you find neighbors until you reach the overlap. This graph-implementation yields a
giant dictionary that maps from compounds to reactions. This exact algorithm is titled ‘Pickle’ within the
code. We discussed sharing the codes more, this will require getting access to Dr. Bhalerao’s SVN server
so that we can place the code in one web repository. The IMPTools server is a development that can
host Django, but it needs Apache. The code ‘SSH’ will allow us to access the server. Some goals for next
week will be to parse the KEGG database using Python. We looked at the abstract some more and also
discussed the Department of Defense competition. May newsletter came out!
"
