Team:Tongji Software/Implementation
Implementation
How to use SynthesisNavigator?
You can carry out the operation of synthetic biology with the metabolic simulation environment on the computer. Users should be clear about demands and then refer to the handbook to input the initial state of molecules. Then DMS will simulate the metabolic environment of E. coli, generate metabolic reactions and molecules, reach the metabolic equilibrium state, and finally give users the changes of molecules. Users can also combine the Pathway Finder to obtain the synthesis pathways and the correlative reaction between molecules.
Who are the end-users?
Our project is prepared for researchers who engaged in synthetic biology or metabolism biology.
How do we implement our project in the real world?
Because our project is a computational simulation, which is different from the real experiment, the results may sometimes not be accurate. Our goal is to provide the experimenter with an intermediate knowledge finder. Users can get the molecular changes through DMS. After the expected changes (which you want) are obtained, the initial state of the simulation can be used for experiments. This is equal to using computer simulations instead of a lot of experiments. At the same time, the changes of molecules inspire users to search for pathways between molecules. After knowing the pathways, users can shut down these reactions by inhibiting specific enzymes, which can close this synthetic pathway. Moreover, if users want to synthesize a target molecule from one molecule, we will provide the computational simulation of this synthesis so that users can avoid too many ineffective experiments. There are so many links to the real experiments that I could not enumerate them here.
What are the safety problems?
Because users run the Synthesis navigator on a computer, there is no danger other than radiation from computers.
What other challenges do we need to consider?
We still need to consider the following issues:
First, initial state is not accuracy.
The initial state of the molecules in the metabolic simulation are given the same value, which is not entirely consistent with the inequality number of different molecules in reality. But this problem can be solved by input special_value with a little more detail, even though it's a hassle for users. In the future, we may provide a script generator. Users can provide CSV or TXT files of molecules and their molecular levels. We use this file as the initial state of molecules or consider making the molecular levels change randomly within a specified size on both sides of default_value to generate an initial state.
Second, common molecules are always in the top rank.
Part of the molecule (such as oxygen, carbon dioxide, etc.) always appears in the top rank because they participate in more reactions. But some users may want to see more of other molecules rank (other users may want to see these common molecular). We can't solve this problem directly; users need to filter Useless data with prior knowledge or input the molecules you want to the observation_list.
Third, complexity of computation.
Our DMS contains a complex computing system, with complicated and broad data. According to the size of the epochs, we may require a great deal of work. In some extreme cases, there may be calculated too long, the memory error or server crash. We wish that users don't input too big a molecular initial state, to avoid such problems.
