Team:CPU CHINA/Model
MODEL SUMMARY
In an iGEM program, modeling is naturally a critical component that make up a extremely significant proportion of the entire work. Our modeling section is devoted to the improvement of essential protein production in the molecular machine and optimize the key enzyme molecular structure and stability. Hence, our modeling section is divided into two parts, mathematical modeling and molecular modeling.
MATHEMATICAL MODELING
The production of two pivotal enzyme proteins, aryl alcohol oxidase (AAO) & manganese peroxidase (MnP), could be imposed on tremendous importance in the industrial application of our molecular machine. Due to the lack of data in the protein production processes, we designed two respective single yeast in vivo kinetics models for both proteins. The crucial stages in vivo, transcription, translation, and mRNA degradation, were all taken into consideration in our system dynamic modeling of several intermediates' & products' concentrations. In the simulation of these concentrations, we utilized Michaelis-Menten equation and law of mass action for the accuracy. While without some of the parameters that we needed, we acquired the similar/exact values in the public database or estimation.
Finally, the results revealed the productions of both enzymes would satisfy our needs and the MnP production could be enhanced through molecular optimization for the balance between AAO production and MnP production. We also proposed a method for predicting the protein production rate which would definitely facilitate the production simulations.
Overall, our mathematical modeling provided our project a criteria for promoting our molecular machine to actual industrial production in the future and paved the way for our molecular optimization and simulation in the molecular part.
MOLECULAR MODELING
Molecular modeling has a significant role in promoting our iGEM project. For our project it was magnificent to improve the stability of MnP protein. So in this section, we simulated the following dimensions based on MnP structure from RCSB database. From RMSD & RMSF indices, we simulated the protein kinetics and examined its stability and screened the amino acid residues for thermal stability improvement. Besides, we focused our attention on the construction of the single-point mutation library and implemented saturated single-point mutations on selected amino acid residues. In order to determine the mutation activity of each mutant, we used machine learning methods to make predictions as well. Moreover, we explored the protein stability of MnP with Rosetta and designed high net charge proteins to improve stability and the hydrophobic gap inside the protein. Meanwhile, we employed quantum chemistry methods to predict the structure of the product obtained by the degradation of PE by MnP. In short, this part of molecular simulation provides some guidance for wet experiments in improving the stability of MnP and presented practical instruction for the design of our molecular machine.
Fig.1 Overview of our modeling section in CPU_CHINA 2021.