Our whole project aims to explore precise approach of preventing plant diseases.

Our wetlab work focuses on validating and improving the efficiency of our biopesticides. We had confirmed the reliability of the key part for its function, including TLS, RdRP and plant immune enhancement. As the presence and transport of RNA vaccines in plants is complex, precise models are needed to provide reliability of our plant RNA vaccines. Our model predicts the diffusion rate and half-life of RNA vaccines in plants, as well as the binding of plant antibodies to pathogen effector. We have also modeled the spread of plant pathogens in natural environments to help put our products into practical use. Our hardware developed a fluorescence detection device to aid in our wet experiments. Moreover, our implementation and entrepreneurship had proved the marketing possibility of our products. It is a rigorous evaluation on the commercial value of the project, indicating the bright future of our product.

Figure 1: An overview to the proof of concept. The picture has shown the interation between each part.

1.1 TLS triggers mRNA transport in plant

Our wetlab work verifies that Met and Ala tRNA sequences from Arabidopsis thaliana, and Met tRNA sequence from tobacco can drive the transport of transcripts within plants when ligated at the 3' end of mRNAs, demonstrating that TLS can triger the movement of RNAs.

Figure 2: GUS staining showing TLS-triggered GUS mRNA transport.

1.2 RdRP

We tried to investigate the gain effect of TMV replicon on target RNA. By expressing target proteins in the TMV subgenome, we demonstrate that different level of gain can be achieved using two subgenomic promoters, and that TMV RNA-dependent RNA polymerase can increase target expression at both protein and RNA levels.

Figure 3: GFP fluorescence signal was significantly enhanced when co-expressed with modified RdRp

1.3 Plant immunity

We successfully modified a natural plant immune receptor (NLR), and the method is to replace its original ID sequence (the original pathogen-recognizing sequence in plant) with a nanobody that specifically recognizes the effector. This modified NLR would improve plant’s ability to recognize pathogenic bacteria while retaining original NLR’s ability to stimulate downstream immune responses, thereby enhancing the resistance of plants to pathogenic bacteria.

1.4 Biosafety

Our wetlab work verifies that two phototoxic proteins, KillerRed and miniSOG, could be expressed in Agrobacterium tumefaciens under the control of the 50Spro promoter. They can play a certain role in controlling the number of engineered bacteria.

To have a whole comprehension of our wet lab work, please turn to RESULTS page.

We combined large eddy simulations of macroscopic plant virus spread and microscopic RNA nonlinear kinematics to validate the whole process of plant RNA vaccines considering the biodynamic behavior of RNA.

We use finite element method to simulate the environmental flow field near the Yuquan campus of Zhejiang University, and the diffusion coefficient in the nearby environment is obtained. Pattern dynamics are used to simulate the spatial spectrum of nonlinear diffusion. Aiming at the problem of particle flow at the microscopic scale, the physical morphology of particles is studied according to the entropy elasticity of the polymer chain and the random walking model, and the Gaussian distribution of the polymer chain under statistical head-to-end distance is derived, and the physical morphology under the electrical action of the Poisson–Boltzmann equation is characterized. Considering the stochastic dynamics model of particles at the microscopic scale, the Lagrange description system is adopted in the fluid, and the corresponding Ito-Langevin equation is established according to the Wiener process. At the macro scale, the equivalent method is used to convert the transpiration tension caused by plant transpiration into Poiseuille flow in the round tube, and a cross-scale model is established by combining the random dynamics of microscopic particles, and the finite element method is used to simulate and verify that the nonlinear dynamics of particles are affected by the size and morphology, flow and vascular pipes of particles. Finally, a simple deep learning model and antigen-antibody binding problem are briefly discussed for this RNA vaccine protocol.

To see the full result of our model, please turn to the MODEL page.

In order to achieve good use effects, we developed the implementation for FloraSentinel, including most of the using points that need to be taken into account. From the production transportation and storage to the correct and effective method of application, we have tried to give our guidelines for how to use it. We also developed the implementation of our software for users to conveniently predict antibody sequences from antigen structures.

To have a full understanding to our user manual, please turn to the IMPLEMENTATION page.