24小时热门版块排行榜

返回列表

当前主题已经存档。

当前只显示满足指定条件的回帖，点击这里查看本话题的所有回帖

allenjobsf853

铜虫 (小有名气)

应助: 1 (幼儿园)
金币: 74.6
散金: 207
帖子: 300
在线: 39.6小时
虫号: 497378
注册: 2008-02-02
性别: GG
专业: 水环境与生态水利

[交流] 【求助】全原子模型和united atom模型的区别？

在论文上看见在进行模拟计算的时候，有些论文说采用全原子模型，而不采用united atom，请教一下，如何在模拟中体现出来使用的是哪种模型？

和同学讨论了一下，有人认为这个是根据使用的力场的不同来区分的，比如compass力场就是全原子模型，而有些力场就不是全原子模型

而我个人为是根据vdw和electrostatic的summation方法选择atom base和group base的不同造成的，如果将基团看做group base，应该就属于united atom了吧？

还是很迷惑，希望高人传道授业解惑下。

先谢谢了

[ Last edited by zdhlover on 2009-12-10 at 15:32 ]

回复此楼

» 收录本帖的淘帖专辑推荐

TOMA'S INDEX

» 猜你喜欢

1楼 2008-12-18 12:40:03

已阅回复此楼关注TA 给TA发消息送TA红花 TA的回帖

wgpig

木虫 (小有名气)

应助: 0 (幼儿园)
金币: 2504.6
帖子: 191
在线: 273.7小时
虫号: 35477
注册: 2004-02-03
性别: GG
专业: 生物大分子结构与功能

★ ★ ★ ★
lei0736(金币+4,VIP+0):谢谢圣诞快乐

United Atom Force Fields and Reduced Representations
In our discussion so far, we have assumed that all of the atoms in the system are explicitly
represented in the model. However, as the number of non-bonded interactions scales with
the square of the number of interaction sites present, there are clear advantages if the
number of interaction sites can be reduced. The simplest way to do this is to subsume
some or all of the atoms (usually just the hydrogen atoms) into the atoms to which they
are bonded. A methyl group would then be modelled as a single 'pseudo-atom' or
'united atom'. The van der Waals and electrostatic parameters would be modified to take
account of the adjoining hydrogen atoms. Considerable computational savings are possible;
for example, if butane is modelled as a four-site model rather than one with twelve atoms
then the van der Waals interaction between two butane molecules involves the calculation
of sixteen terms rather than 144. Other hydrocarbons are often represented using united
atom models. Many of the earliest calculations on proteins used united atom representa-
representations In this case, not all of the hydrogen atoms in the protein are subsumed into their adja-
adjacent atoms, but just those that are bonded to carbon atoms. Hydrogen atoms bonded to polar
atoms such as nitrogen and oxygen are able to participate in hydrogen-bonding interactions,
which are modelled much better if these hydrogens are explicitly represented.
One drawback with a united atom force field is that chiral centres may be able to invert
during a calculation. This was found to be a problem with the united atom force fields for
proteins. The alpha carbon in the peptide unit (Ca in Figure 4.42) is bonded to a hydrogen
atom and to the side chain (glycine and proline are slightly different; see Section 10.1). A
united atom force field model would not explicitly include the alpha hydrogen. Unfortu-
Unfortunately, the stereochemistry at the alpha carbon can then invert during a calculation. This
should be avoided as the naturally occurring amino acids have a defined stereochemistry
(as shown in Figure 4.42). This inversion may be prevented through the use of an improper
torsion term (e.g. N-C—Ca-R) to keep the side chain in the correct relative position.
In a united atom force field the van der Waals centre of the united atom is usually associated
with the position of the heavy (i.e. non-hydrogen) atom. Thus, for a united CH3 or CH2
group the van der Waals centre would be located at the carbon atom. It would be more
accurate to associate the van der Waals centre with a position that was offset slightly
from the carbon position, in order to reflect the presence of the hydrogen atoms. Toxvaerd
has developed such a model that gives superior performance for alkanes than do the simple
united atom models, particularly for simulations at high pressures [Toxvaerd 1990]. In
Toxvaerd's model the interaction sites are located at the geometrical centres of the CH2 or
CH3 groups. The forces between these sites act on the united atom mass centre, which
remains located on the carbon atom (with a mass of 14 for a CH2 group and 15 for a CHj
group). As the interaction site is no longer located at an atomic nucleus the forces acting
on the masses are more complicated to calculate, but little additional computational expense
is required. The effect of using such an anisotropic potential is nicely illustrated by the two
arrangements of methylene units shown schematically in Figure 4.43. In the united atom
model both arrangements would have the same energies and forces, but this is not so
with the Toxvaerd anisotropic potential.