Modeling is the bridge between science and engineering in this project; using the scientific knowledge on our system, we can describe how the system will act and interact. Furthermore, using the quantitative relations derived from the models the circuit has been adjusted and redesigned. This description is used for predicting what will happen in the lab, but also if the project can be applied in reality in a 'band-aid' product. Next to giving a preliminary design of the band-aid, new peptides are designed to give us the properties we want for this application.
By modelling the behavior of the circuit, we are finding answers for the following questions:
- How many minutes does the cell lysis take from the point of induction?
- How much peptides are produced by the circuit?
- How much peptides are still uncleaved by the SUMO at the point of cell lysis?
In order to simplify this modeling, we split it up into three different modules which were used as building blocks for the final model. In this way the feasibility of the total model is more easily validated, as the building blocks can be validated independently. The three building blocks are:
In the total circuit model the questions stated above are answered. Moreover, to indicate the validity of these answers, we perform a sensitivity analysis to investigate which parameters most influence the results. This gives an indication of the validity as some parameters are quite well known well others are less sure.
Application modeling (band-aid)
One of the most important parts of the project is to indicate the final application. For that reason, band aid modelling was performed. To be more specific, the idea of the final application is related to placing our engineered organism into a band aid.The band aid will be located on the wound. In that way, MRSA will be detected and the peptides will be produced and released through the band aid. By modelling the band aid, we are trying to answer questions like:
- If the MRSA is detected, is the amount of the produced peptide enough to kill it ?
- How many pores are necessary in order to be possible for the peptide to be released?
We also decided to design novel peptides which meet our specific needs: very toxic for S.aureus (very low MIC), not toxic for humans and E.coli. These properties we also had in mind by choosing from the existing peptides, but the matching peptides did not met the requirements that well. In order to manage this, we performed data analysis and feature extraction of the existing experimentally validated datasets combined with association rule mining.