What did our model achieve?

We achieved 4 main aims in our modelling work:
1、We simulated the whole process from plasmid introduction to protein degradation,which deepens our understanding of reaction mechanism. The real experimental data is also used to improve our model, with the help of network mechanism analysis and parameter relationship analysis.

2、We used both least squares and neural networks to analyze the data and obtain the relationship between protein degradation and plasmid concentration,the use of which two methods makes the entire construction more reliable and provide insights for other teams who want to use this approach.

3、We offered some suggestions to wetlab according to sensitivity analysis,helping them to find the reason on how different factors work.

4、We detected the appropriate check point to find the time of protein degradation to a certain proportion.By this means,we could offer advice on the amount of plasmid introduction, combined with patients’needs to protein degradation time and concentration of protein at equilibrium during future practice .

What were we modelling?

We focused on building ODE models to describe the whole process from plasmid introduction to protein degradation ,predicting different circumstances observed in wetlab. Based on these minds,we model two different antibodies GFP and ErbB3 respectively. We simulated the two process of introducing plasmids carrying different antibodies （GFP and ErbB3）into cells respectively. The majority of our experimental data came from the proof-of-concept study of the GFP system,as the reaction network of the two systems is the same and only some kinetic parameters differ.

How did we model?

Model parameterization: we modeled the whole process of our system to mathematize the process, using kinetic and dynamic models solved by analytical and numerical simulation techniques.（Law of mass action ,antigen-antibody reaction kinetics and our own equations ）

Computer assistance:we used Simbiology for solving differential equations, data visualization, sensitivity analysis, parameter sweeps and comparison with existing data to verify the accuracy of the model.

Data analysis:we used both least squares and neural networks to analyze the data and obtain the relationship between protein degradation and plasmid concentration,the use of which two methods makes the entire construction more reliable.

Donec id elit non mi porta gravida at eget metus. Fusce dapibus, tellus ac cursus commodo, tortor mauris condimentum nibh, ut fermentum massa justo sit amet risus. Etiam porta sem malesuada magna mollis euismod. Donec sed odio dui.