Classical Molecular Dynamics (MD) simulations with all-atom Force Fields (FF) represent an extremely powerful tool for the "in silico" investigation of molecular bio-systems, that was developed and optimized in the last three decades and has reached a noticeable accuracy. However, the microsecond time-scale for systems as large as protein is reachable only with extensive parallelization, limiting the application of this approach to short steps of the biological processes. In fact, those occurr typically much larger time and size scales. A number of low resolution simplified models was recently developed in order to overcome this limitation: through the reduction of degrees of freedom it is possible to treat the biologically relevant scales. Minimalist models for proteins and nucleic acids are presented here. The one-bead per amino acid/nucleotide coarse graining (CG) level is chosen, since it is the highest level of CG that still allows the explicit description of the secondary structure transitions. This extreme simplification is paid by additional difficulties in the parameterization of the FF. In spite of this, provided proper functional forms for the FF terms are chosen, it is still possible to obtain quantitatively predictive simulations. Results on the secondary structure transitions in homo-polypeptides (helix-sheet) and in DNA (A to B DNA and denaturation) are reported. The simulations show a realistic dynamics and thermodynamics and a noticeably good comparison with experimental data. Fine effects, such as the formation of denaturation bubbles in DNA, are also observed in the simulations. In conclusion, minimalist but extremely flexible models for polypeptdides and nucleic acids are presented, that are capable of predictively describe structural transitions in biological polymers. Work in progress include the optimization of the FF in a amino acid (or nucleotide) specific fashion to improve its transferability. This will open to a large variety of applications.
PHYS 138-One-bead models for structural transitions in macrobiomolecule
Tozzini;Valentina
2008
Abstract
Classical Molecular Dynamics (MD) simulations with all-atom Force Fields (FF) represent an extremely powerful tool for the "in silico" investigation of molecular bio-systems, that was developed and optimized in the last three decades and has reached a noticeable accuracy. However, the microsecond time-scale for systems as large as protein is reachable only with extensive parallelization, limiting the application of this approach to short steps of the biological processes. In fact, those occurr typically much larger time and size scales. A number of low resolution simplified models was recently developed in order to overcome this limitation: through the reduction of degrees of freedom it is possible to treat the biologically relevant scales. Minimalist models for proteins and nucleic acids are presented here. The one-bead per amino acid/nucleotide coarse graining (CG) level is chosen, since it is the highest level of CG that still allows the explicit description of the secondary structure transitions. This extreme simplification is paid by additional difficulties in the parameterization of the FF. In spite of this, provided proper functional forms for the FF terms are chosen, it is still possible to obtain quantitatively predictive simulations. Results on the secondary structure transitions in homo-polypeptides (helix-sheet) and in DNA (A to B DNA and denaturation) are reported. The simulations show a realistic dynamics and thermodynamics and a noticeably good comparison with experimental data. Fine effects, such as the formation of denaturation bubbles in DNA, are also observed in the simulations. In conclusion, minimalist but extremely flexible models for polypeptdides and nucleic acids are presented, that are capable of predictively describe structural transitions in biological polymers. Work in progress include the optimization of the FF in a amino acid (or nucleotide) specific fashion to improve its transferability. This will open to a large variety of applications.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
