Team:Illinois-Tools/Notebook/Week6


 * back to notebook

<>

6/28-7/4
Donny

At this time, I have a working prototype for the site. It has a user interfaces with forms and/or dropdown menus to select the desired compound. From there the site will return the shortest path (by number of reactions). It displayed the KeggID and common name of each intermediate, as well as the reaction ID's that correspond to the returned pathway. It's final feature to date, is listing the cofactors and additional substrates that were assumed to be present for the pathway to succeed.

Kanishka

Biobricks!! - The goal of our project will be to provide the user with all the biobricks that code for all the enzymes needed for a particular reaction pathway. So first, we will have to figure out the genes that code for all the enzymes that will carry out a particular pathway. First, we will take the nucleotide sequences for these genes and check whether the exact sequence already exists in the existing biobrick database. So we will have to scan through the biobrick registry to compare our sequences with the existing sequences. Since the current biobrick registry only has about 3200 parts, it is highly unlikely that the exact sequence will already be existing. However, if any of the sequences do already exist, the program will tell the user that the biobrick for that particular gene already exists, and for the remaining sequences, the program will construct biobricks for all these genes and display them for the user. Now from what I think, each biobrick will be one gene. (Because one enzyme cant come from two different genes, but one gene can code for several proteins). Each gene can either code for a single enzyme, or multiple enzymes. So if a particular reaction pathway contains 20 enzymes, there may be up to 20 different biobricks. The program will add the appropriate suffix and prefix to the sequences, assign them to a biobrick standard, and then present them in a proper biobrick format. (Now Im not sure if we could combine these biobricks into one biobrick, since all the biobricks code for one reaction). So the user then can order all these biobricks, and then in lab, can insert these plasmids into a bacteria and carry out the particular reaction. If we can, we should actually test this out in lab with the help of the wetlab team.

Riyad Worked with Kanishka on biobricks. He pretty much did an excellent job summing it up so not much to add. Right now we are looking into incorporating sometype of BLAST into our program to compare our output gene sequence with the biobrick registry.

Palak

This week I began working on an algorithm that would search for the path with the fewest changes to the host organism. I got the algorithm to function properly but the database did not have all the organisms paired up with the reactions yet. The next steps will be to populate the database with the additional information such as reversibility of reactions and stoichiometry.