Development of Combined Computational and Experimental Approaches for Using Molecular Engineering in the Design, Construction, and Analysis of Integrated Biosensor Microsystems

摘要

Structure-based protein design methods constitute the basis for this project. These methods aim to describe molecular recognition using semi- empirical potential functions that capture van der Waals, hydrogen-bonding, electrostatic, and solvation contributions. These descriptions are combined with representations of the dominant degrees of freedom in a protein design calculations: the sequence and structure of amino acid side-chains placed within the three-dimensional frame work of a parent protein ('the scaffold'), and the translations/rotational degrees of freedom of a ligand. Discrete combinatorial optimization algorithms are then used to identify a combination of amino acids and docked ligand conformation that represents the global energy minimum of the potential function. Computationally generated solutions are then tested experimentally by making the specified mutations in the gene encoding the parent protein using oligonucleotide-directed mutagenesis, producing the mutant protein by heterologous over-expression in E. coli followed by protein purification and appropriate biochemical assays to test for the presence of the desired function.