| Popis: |
Kim, Straub, and Keyes introduced the statistical temperature molecular dynamics (STMD) algorithm to overcome broken ergodicity by sampling a non-Boltzmann flat energy histogram as noted in Kim, Straub, and Keyes, Phys. Rev. Lett. 97: 050601 (2007). Canonical averages are calculated via reweighting to the desired temperature. While STMD is promising, its application has been almost entirely to simple or model systems. In this dissertation the implementation of STMD into the biosimulation package CHARMM is used to simulate the methionine enkephalin pentamer peptide with a methione terminal cap in a droplet of CHARMM TIP3P water molecules. Chain thermodynamics is analyzed from the novel perspective of the statistical temperature as a function of potential energy, $TS(U), automatically generated by STMD. Both the minimum in the slope of $TS(U), and the peak in the heat capacity as a function of temperature, calculated via reweighting, indicate a collapse transition at Tθ ≈ 253K. Distributions of dihedral angles are obtained as a function of temperature. Rotamer regions found in the literature are reproduced, along with unique regions not found previously, including with advanced algorithms, indicating the power of STMD enhanced sampling. |
