Our novel RESCUE system is dependent on computational modeling to bring out insights to facilitate laboratory work and design. Experimental obstacles can be facilitated, and new insights are achieved with adequate support from in silico models. In general, three different approaches to studying the system were ambitiously performed to resolve potential problems during the experimentation phase.
Team:NUS Singapore-Sci/Model
Modelling
Overview
Firstly, the functionality of the fusion protein is the largest uncertainty for the experiment. Protein threading was used to generate its 3D structure based on the amino acid sequence before an attempt using molecular dynamics to bring the molecule to a stable energy state was performed. The intended outcome of this model will be to provide an idea if a functional native state of the protein construct exists.
Secondly, least-square regression (LSR) algorithm was coded in the farsighted view that the design of gRNA may affect the catalytic efficiency of our fusion protein. Patterns of the gRNA design used in the experiments can be picked up through their catalytic efficiencies to prevent a similar gRNA sequence to be redesigned. This may help to reduce unfavorable readings if a wide array of reporter systems are being dealt with in the future.
Lastly, ensemble kalman filter (EnKF) was adapted from the Earth-sciences field to be applied to our experiment. Since EnKF is primarily used to resolve dynamics events with difficult-to-measure parameters and their large associated uncertainties, the same can be applied to resolve the uncertainties and a large set of unknown molecular concentrations within the kinetics model for our RESCUE system, especially the concentration of edited mRNA, which is a direct measurement of editing efficiency. This would allow the laboratory team to infer certain molecular concentrations prior to performing validation tests.
In general, our models are currently proof of concepts, either in debug, or pending results from the experiments. When the models are implemented in the near future, they will be able to facilitate the laboratory team by providing interesting insights especially during the experimentation phase..