注册 登录  
 加关注
查看详情
   显示下一条  |  关闭
温馨提示!由于新浪微博认证机制调整,您的新浪微博帐号绑定已过期,请重新绑定!立即重新绑定新浪微博》  |  关闭

计算生物学实验室

http://www.bioms.net

 
 
 

日志

 
 

Martini ForceField and Coarse Grained Simulation  

2011-10-24 22:13:10|  分类: Gromacs |  标签: |举报 |字号 订阅

  下载LOFTER 我的照片书  |

I. Limitations which should be attentive before you use martini CGFF
1.Temperature range for applications is 270-330K.
As is true for any forecefield, applications outside the temperature range used for parameterization (270-330K) have to be considered with care.


2.The potential range of applications of the CG model is only suitable for fluid phase. Now the CGFF just focus on the protein and lipid. Other soft material should be parameterized by yourself.


3. Electrostatic interaction is not summarized completely.
Apart from the implicit screening in the CG model, the neglect of long-range electrostatic forces poses a further limitation. Pairwise interactions beyond 1.2 nm (between two and three CG beads away) are not taken into account.


4.Finally, in applications of peptides and proteins one has to be aware that secondary structure transformations are not modeled in the current parameterization.


II. Simulation Method
It's very same as Gromacs. If you can play Gromacs, the CG simulation with Gromacs is a piece of cake. There are many tutorials, *.itp files and examples in the following website, you can download and repeat them:http://md.chem.rug.nl/~marrink/coarsegrain.html
However, there also are some difference with Gromacs:


1. CG simulation needs a new command called g_fg2cg, which can transform *.gro/*.trr files between FG and CG. if you want to get it, Your must compile whole special Gromacs because I wanted to compile the original command file instead of whole Gromacs but failed. Actually, you needn't to make the trouble, your can download it executable file from here. However, I don't promise you can use it normally because I compile it under Cygwin system. Additionally, if you can't download it, please mail me.


2. You should try the parameter file (*.mdp) in the examples firstly if you don't familiar with CG simulation.


3. You should focus on the minimization and work hard on all troubles. If failed, you maybe try cg minimization or double precision Gromacs.


III. Some Useful tools for CG simulation


1. pdb2CGpdb-2.1 for preprocess
input file: *.pdb
output file: pdb-CG.pdb (Please keep this file name for being read by the topol-CG.exe!)
./pdb2CGpdb < *.pdb > pdb-CG.pdb


Attentions: Important Notes for Correct Functioning
1. The pdb file should be cleaned of hydrogen atoms, if any, HEADER/REMARK etc, and Nterm and Cterm atoms;
2. The first line of the pdb file must contain the number of residues and the number of atoms;
3. The pdb file should not contain a CHAIN-ID (between the resname and the resnumber);
4. If it indicates that there is wrong with an amino acid, please check the name of all atoms, such as the C atom's name in Ile "CD" should change into "CD1".
5. The pdb-CG.pdb can be revised into *.gro for CG simulation comparing to the fine particles gro file. Here, I advise you used the atom's name as "CA" and "SC".


2. topol-CG-2.1 for preprocess
# input file: pdb-CG.pdb
# output file: protein.itp
The protein.itp can be revised into *.top including itp file for CG simulation. However, the protein.itp is not good, the better itp file is produced from seq2itp command.


3. do-order parameter for analyzing the order parameter
./do-order.sh 0 10000 20 0 0 1 64
# 0: beginning step;
# 10000: 10-ns trajectory;
# 20: average over 20 equally-spaced snapshots;
# 0 0 1: order-parameter(P2) is calculated relative to the Z-axis;
# 64: the half number of lipids in your simulation.
Maybe, you will revise some codes according to your Gromacs installation.
By using this command, You'll get all order-parameters of each bonds in lipid and their average values.


4. seq2itp command for obtaining the correct itp file.
This command needs two files which contains the secondary structure and amino acid sequence. The detail parameters check its help (seq2ipt --help).Here, I also advise you used the atom's name as "CA" and "SC". if so, you will get more beautiful images than default.


5. g_msd for calculating the lateral diffusion after remove jumps over the box boundaries (trjconv -pbc nojump)


6. g_density for getting the bilayer thickness.

  评论这张
 
阅读(1424)| 评论(0)
推荐 转载

历史上的今天

在LOFTER的更多文章

评论

<#--最新日志,群博日志--> <#--推荐日志--> <#--引用记录--> <#--博主推荐--> <#--随机阅读--> <#--首页推荐--> <#--历史上的今天--> <#--被推荐日志--> <#--上一篇,下一篇--> <#-- 热度 --> <#-- 网易新闻广告 --> <#--右边模块结构--> <#--评论模块结构--> <#--引用模块结构--> <#--博主发起的投票-->
 
 
 
 
 
 
 
 
 
 
 
 
 
 

页脚

网易公司版权所有 ©1997-2018