2. MFT design
A Monte Carlo / MFT approach to protein sequence design
2.1 Inverse protein folding problem ?
The inverse folding problem was originally defined by Drexler [1] and Pabo [2] as the problem of defining the sequences compatible with a given protein fold. It is fundamental to protein design and engineering, and, as such, as attracted considerable interest [3-8]. Since the function of a protein is directly related to its three dimensional structure, manipulation of the structure via the sequence changes can provide functional diversity. Protein molecules can be engineered to optimize their activities, as well as to alter their pharmacokinetic properties.
This page describes the method we developed for solving this problem.
2.2 Background
The aim of protein sequence design is to generate sequences compatible with the target structure, but incompatible with competing folds. As such, there is one theoretical issue, and one computational problem to be solved. Here we provide links to general reviews that cover both subjects:
2.3 Our Strategy for Protein Sequence Design
Our protein sequence design uses a Monte Carlo search technique in sequence space, a SCMF technique in structure space, and the Random Energy Model for reaching specificity. [9]
Schematic description of our Monte Carlo protein sequence design procedure. Starting from a random sequence, S0, of the required composition, a model structure is built,
and its energy is evaluated and stored as E0. Two positions are then chosen at random, and the corresponding amino acid types in S0 are exchanged, yielding a new sequence S1.
A new model structure is built based on S1, and its energy is stored as E1. The sequence move from S0 to S1 is accepted or rejected according to the Metropolis Monte Carlo probability The procedure is repeated until the system has equilibrated and the energy remains steady.
and its energy is evaluated and stored as E0. Two positions are then chosen at random, and the corresponding amino acid types in S0 are exchanged, yielding a new sequence S1.
A new model structure is built based on S1, and its energy is stored as E1. The sequence move from S0 to S1 is accepted or rejected according to the Metropolis Monte Carlo probability The procedure is repeated until the system has equilibrated and the energy remains steady.
References
1. Drexler, KE. Molecular engineering: an approach to the development of general capabilities for molecular manipulation. Proc. Natl. Acad. Sci. (USA), 78, 5275-5278 (1981).
2. Pabo, C. Designing proteins and peptides. Nature, 301, 200 (1983).
3. Mutter, M and Tuchscherer, G. Nonnative Architectures In Protein Design and Mimicry. Cellular and Molecular Life Sciences, 53, 851-863 (1997).
4. Smith, CK and Regan, L. Construction and Design Of Beta-Sheets. Accounts Of Chemical Research, 30, 153-161 (1997).
5. Cao, AN, Lai, LH and Tang, YQ. The Current State and Prospect Of De-Novo Protein Design. Progress In Biochemistry and Biophysics, 25, 197-201 (1998).
6. Giver, L and Arnold, FH. Combinatorial Protein Design By In-Vitro Recombination. Current Opinion In Chemical Biology, 2, 335-338 (1998).
7. Regan, L and Wells, J. Engineering and Design. Recent adventures in molecular design. Curr. Opin. Struct. Biol., 8, 441-442 (1998).
8. Shakhnovich, EI. Protein Design : a Perspective From Simple Tractable Models. Folding & Design, 3, R45-R58 (1998).
9. Koehl, P and Levitt, M. De novo protein design. I. In search of stability and specificity. Journal of Molecular Biology, 293, 1161-1181 (1999).