In silico 3D-modeling and Molecular dynamic analysis ofTGFBI and its variants:
The nucleotide sequence of TGFBI protein was retrieved from NCBI (ID), and the structure was modeled through comparative modeling using the Robetta server (Song et al. 2013).The structure of full-lengthTGFBI was modeled using the coordinates of PDBID:5NV6(Garcia-Castellanos et al. 2017). For further refinement, the model was subjected to molecular dynamics (MD) simulation for 200ns. The refined structure was validated using MolProbablity Server(Williams et al. 2018). The last frame of the 200ns simulation was used as a template for modeling the mutant forms of TGFBI, including A323E, A323V, and 649 frameshift. The modeled mutants were then simulated for 200ns each in order to understand the impact of variations on the structure and dynamics of the TGFBI protein.
MD simulations of the modeled proteins were done employing GROMACS software version 2021.5(Van Der Spoel et al. 2005). CharmM36 forcefield was used for generating the topology, and the system was solvated using the TIP3P water model(Vanommeslaeghe et al. 2010). Neutralization of the system was done by the addition of the appropriate number of sodium ions. Equilibration of the system was then performed under successive NVT and NPT ensembles for 500ps each. The production simulation was run using a leap-frog dynamics integrator and a step size of 2fs with the consideration of periodic boundary conditions in all three dimensions. The trajectory was analyzed. Principal component (PC) analysis was performed to understand the global motions of the TGFBI and the three mutant forms. The covariance matrix was generated from the Cα atom coordinates and diagonalized to obtain the eigenvectors. Since the first two eigenvectors described the major proportion of the trace of the covariance matrix, the trajectory was projected along the first two principal components. Free energy landscape was further generated using the 2-dimensional projection of the trajectory along PC1 and PC2.