Team:NUDT CHINA/Model/Discussion

Designed Protein Degradation Method Based on

Trim21 And Nanobody              -- Discussion


Future Work

From the previous analysis, it can be seen that the idea based on increasing the concentration of the input plasmid by increasing the rate of protein degradation or reduce the concentration of protein is limited by the intracellular XXX , and increasing plasmid dosage also has a non-negligible limiting effect on cell growth and proliferation. Therefore, we hope to explore how to complete the entire circuit through modeling, and improve the threshold of the reaction speed from the underlying improved circuit. When the original constructed plasmid is expressed, the translated trim21 and the antibody peptide chain are cleaved by the p2a method, and the function of trim21-mediated ubiquitination functions after the antibody-antigen complex (Fig.1).


Fig.1 This is the original design.

However, if the antibody and trim21 are directly linked after expression by designing the gene sequence, which means that trim21 is also attached to the complex (Fig.2).


Fig.2 This is the hypothetical design.

Therefore, Trim21 can directly initiate the ubiquitination of the antigen-antibody complex.(It is reflected in the equation of the model to skip the S + E3 <-> SE3 link) . Through computer simulation,we evaluate the effect of this improvement and compare it with the original design. (Fig.3)


Fig.3.1 This is the comparison of simulation results between the two designs.


Fig.3.2 This is the comparison of simulation results between the two designs.

In the process of ubiquitination, if the ub molecule on the ubiquitin chain is considered, it will randomly fall off and eventually decompose into ub and S.However, plasmids only express the trim- antibody complex, without a separate trim, cannot bind S, cannot complete the reubiquitination of S, and finally part of S cannot be decomposed.

Therefore, we further envisioned that part of the original design plasmid was introduced to solve the problem at the same time, and the import proportion could be simulated by computer.


1、Sub-module modeling is a practical way to abstract practical problems. Modularity is like building blocks,as blocks can be transformed into different shapes through different combinations. After the first design, the next time you only need to combine the module unit. In order to create a useful model that can describe the circuit designed by our team, we split it into three modules (plasmid expression process, which has a broad application base in modeling; antigen-antibody binding to form a complex process, based on antigen-antibody reaction Trim-mediated complex ubiquitination to the final degradation process, based on enzymatic reactions and combined with the law of mass action)modeled separately, and finally integrated. This design facilitates the establishment of the model and provides information for subsequent team modeling.

2、The computer is used to solve the parameters and complete the model. The model we built is complex, with a large number of equations and parameters, but it can be efficiently computed using a computer which is the engine of the entire model. We use two methods(least squares and neural networks)to process the data from the wet group and enhance the usable value of the model. And the related implementation of the two methods has been uploaded to Analysis, providing guidance to teams that wish to use this method in the future. Finally, we take some data as a test group to test the reliability of the model.

3、As for our analysis, we aim to provide insights for wetlab and future applications. Through model, the whole process of circuit operation is explicitly expressed, which helps the wet group to further understand the reaction mechanism. We use MATLAB to visually demonstrate the weight of each reactant's influence on the experiment, give suggestions for checkpoint selection, and propose an improved circuit design to fundamentally improve the efficiency of the system.

4、Finally, limited by the laboratory conditions and the amount of data and the accuracy of the equation description process, it is unavoidable that the values of the parameters and the model's description of the system are biased. However,the more important value in this model lies in the idea of modeling: modular modeling, multi-method co-fitting, through sensitivity analysis and parameter sweeps to get reasonable and powerful conclusions, which can be applied to other model constructions as a programmatic analysis tool.