| 查看: 3653 | 回复: 6 | |||
[交流]
关于用amber跑REMD计算的流程及输入文件的设置
|
|||
|
最近看了一篇论文，http://www.ncbi.nlm.nih.gov/pubmed/22339436 作者使用amber做了REMD（replica exchange molecular dynamic）的计算。 自己正好也有类似需要，想用REMD模拟。就想按论文相同的条件实现一下，不过论文里的过程写的不是很详细。下面是原文的method部分： In this work, the temperature setting used for sheep PrP 125-230 such that MD simulations were exchanged at 320.0, 322.0,324.0, 326.0, 328.1, 330.2, 332.3, 334.4, 336.5, 338.6, 340.8, 343.0, 345.2, 347.4, 349.6, 351.8, 354.0, 356.2, 358.5, 360.8, 363.1, 365.4, 367.7, and 370.0 K was used. All of the replicas were equilibrated for 20 ns without exchanging temperatures and then extended for 65 ns of REMD simulation. The generalized Born model used in this study modified the calculation of Born radii and improved the accuracy in the solvent polarization for macromolecules. The combinational use of the all-atom point-charge force-field (also known as ff03) and the generalized Born model led to successful folding of several proteins. The AMBER 11 simulation package 26 was used in both REMD simulation and data analysis. The melted huPrP 121-230 was computed starting from an extended huPrP. To generate the initial extended structure, a heating method was applied to a known NMR structure (PDBcode 1hjn,15Figure 1A), enabling it to unfold at 600 K for 40 ns of MD simulation to result in an extended conformation (Figure 1B) as described previously. During this simulation, the disulfide covalent bond between residues 179 and 214 was preserved. In total, 24 replicas with duration of 65 ns and with an integration time step of 2 fs were computed based on the extended huPrP with different random number seeds to generate the initial conditions. A 16 Å force-shifted non-bonded cutoff and generalized Born solvent models with salt concentration of 0.2 M were applied. 根据文章的描述我自己总结了一下，并按文章描述写了各个环节的输入文件。想和大家交流一下，看看我写的有没有不合适的问题。 Simulation Procedure: 1. system building 2. system minimization 3. heating system 4. generate the extend conformation 5. local minimization after heating system 6. equilibrate the every replica 7. REMD simulation 1.leap.inp for system building source pdb: 1ag2 use the the ff03 (Duan et al.) force field leap.inp source leaprc.ff03.r1 # load pdb file 1ag2 = loadPdb input.pdb # solvation # solvatebox 1ag2 TIP3PBOX 18.0 # save 1ag2 to pdb file savePdb 1ag2 1ag2.pdb # add countinos # addions 1ag2 Cl- 0 # addions 1ag2 Na+ # s-s bond bond 1ag2.179.SG 1ag2.214.SG # save 1ag2 to prmtop and inpcrd files saveAmberParm 1ag2 1ag2.prmtop 1ag2.inpcrd # finish quit 2. system minimization minimisation for heated system &cntrl imin=1, maxcyc=1000, ncyc=500, cut=999., rgbmax=999.,igb=1, ntb=0, ntpr=100 / ~ ~ 3. heat the system heating system from 0 K to 600K. &cntrl nstlim = 50000, dt = 0.002, ntt = 1, tautp = 1.0, tempi = 0, temp0 = 600, ntc =2, ntf = 2, ntpr =100, ntwx = 100, ntb = 0, igb = 1, #这里设置和后面remd的设置不同（igb = 5）结果可能受影响。 cut = 999.0,rgbmax = 999.0, / 4. generate the full unfolded conformation To generate the initial extend structure, a heating method was used to a known NMR structure (PDB code:1ag2), enabling it to unfold at 600 K for 40 ns of LD simulation to result in an extended conformation. 40nsld.inp enabling the heated NMR structure to unfold at 600 K for 40ns of LD simulation &cntrl irest = 1, ntx = 5, nstlim = 20000000, dt = 0.002, ntt = 3, gamma_ln = 1.0, tempi = 600,temp0 = 600, ntb = 0, igb = 2, ntpr = 500, ntwx = 1000, ntwr = 2000000, ntc = 2, ntf = 2, cut = 999.0,rgbmax = 999.0, / | RE_POSITION Moving by -2.485812 -1.096141 0.618430 NSTEP = 20000000 TIME(PS) = 40060.000 TEMP(K) = 577.57 PRESS = 0.0 Etot = 1351.4737 EKtot = 2407.9819 EPtot = -1056.5082 BOND = 626.4085 ANGLE = 1429.4823 DIHED = 1419.4285 1-4 NB = 376.4803 1-4 EEL = 3772.0639 VDWAALS = -368.8173 EELEC = -5949.1848 EGB = -2362.3695 RESTRAINT = 0.0000 ------------------------------------------------------------------------- A V E R A G E S O V E R ******* S T E P S NSTEP = 20000000 TIME(PS) = 40060.000 TEMP(K) = 600.20 PRESS = 0.0 Etot = 1514.3153 EKtot = 2502.2899 EPtot = -987.9746 BOND = 644.7390 ANGLE = 1506.2954 DIHED = 1362.4635 1-4 NB = 397.8849 1-4 EEL = 3727.1078 VDWAALS = -361.6498 EELEC = -5967.0181 EGB = -2297.7973 RESTRAINT = 0.0000 ------------------------------------------------------------------------- 结果Eptot 的plot图 600kmd.inp enabling the heated NMR structure to unfold at 600 K for 40ns of LD simulation &cntrl irest = 0, ntx = 1, nstlim = 20000000, dt = 0.002, ntt = 3, gamma_ln = 1.0, tempi = 600,temp0 = 600, ntb = 0, igb = 2, ntpr = 500, ntwx = 1000, ntwr = 2000000, ntc = 2, ntf = 2, cut = 999.0,rgbmax = 999.0, / 600kmd.out ... | RE_POSITION Moving by -0.033675 -0.790281 -1.085325 NSTEP = 2000000 TIME(PS) = 4060.000 TEMP(K) = 601.00 PRESS = 0.0 Etot = 1554.3820 EKtot = 2505.6360 EPtot = -951.2540 BOND = 654.9106 ANGLE = 1485.1469 DIHED = 1383.1021 1-4 NB = 422.1188 1-4 EEL = 3774.2620 VDWAALS = -379.3759 EELEC = -6038.4682 EGB = -2252.9505 RESTRAINT = 0.0000 ------------------------------------------------------------------------- A V E R A G E S O V E R 2000000 S T E P S NSTEP = 2000000 TIME(PS) = 4060.000 TEMP(K) = 599.90 PRESS = 0.0 Etot = 1513.2011 EKtot = 2501.0782 EPtot = -987.8771 BOND = 644.5795 ANGLE = 1506.1933 DIHED = 1361.8286 1-4 NB = 398.1118 1-4 EEL = 3724.2906 VDWAALS = -363.4705 EELEC = -5968.1047 EGB = -2291.3058 RESTRAINT = 0.0000 irest = 1, ntx = 5,和irest = 0, ntx = 1 模拟结过差别还是有的，从EPtot的结果来看，后者结构变化比较激烈，前者较稳定。但是不知道在这个模拟里用哪个合适？ 5. local minimization after heating system 使用第三步minimization的输入文件。 6. equilibrate the every replica equilibrate.mdin equilibration 20 ns, every 10ps save output. equilibration &cntrl irest=0, ntx=1, nstlim=10000000, dt=0.002, irest=0, ntt=3, gamma_ln=1.0, temp0=XXXXX, ig=RANDOM_NUMBER, ntc=2, ntf=2, nscm=1000, ntb=0, igb=5, cut=999.0, rgbmax=999.0, ntpr=5000, ntwx=5000, ntwr=10000000, nmropt=1, / &wt TYPE='END' / DISANG=system_chir.dat 7. REMD simulation remd 40ns exchange every 2ps <- 这里一直有困惑，到底间隔多久交换一下比较合适？ remd.mdin remd 40ns exchange every 2ps &cntrl irest=0, ntx=1, nstlim=1000, dt=0.002, irest=0, ntt=3, gamma_ln=1.0, temp0=XXXXX, ig=RANDOM_NUMBER, ntc=2, ntf=2, nscm=1000, ntb=0, igb=5, cut=999.0, rgbmax=999.0, ntpr=100, ntwx=1000, ntwr=100000, nmropt=1, numexchg=20000, / &wt TYPE='END' / DISANG=system_chir.dat |
回复此楼» 本帖已获得的红花（最新10朵）
超人与小木虫» 猜你喜欢
科研人应该花精力去思考如何解决问题，而不是去凝练问题
已经有12人回复
-
要骂人了，新模版改版就是要淡化问题凝练这种虚的东西，结果有个评委还在说凝练得不够
已经有17人回复
-
基金系统什么内容也没有
已经有10人回复
-
学科评审组评审是指会评吗？
已经有5人回复
-
梦想
已经有8人回复
-
面上合作单位盖章
已经有9人回复
-
面上函评意见出来了，像什么等级？
已经有17人回复
-
为什么资助数各大高校都创新高，自己申请怎么就这么难
已经有14人回复
-
小白求助 投论文要求的highlights应该如何写
已经有4人回复
-
麻烦专家们看看评委们的意见（F口面上）
已经有12人回复
» 本主题相关价值贴推荐，对您同样有帮助:
- amber跑蛋白复合物动力学前，小分子用高斯如何处理
已经有7人回复
- AMBER MMPBSA 计算时建立prmtop文件问题
已经有6人回复
- 新手想知道如何用Amber做蛋白质分子固定某两个或多个氨基酸距离的MD计算
已经有4人回复
- 【转帖】用Amber+Gaussian做小分子力场
已经有16人回复
» 抢金币啦！回帖就可以得到:
-
云南大学丁军桥团队2027年博士招生
+1/809
-
【征稿中！】2026年机械、制造技术与材料国际会议 (MEMTM 2026)
+1/768
-
2027级博士研究生——质谱分析
+5/125
-
广西大学作物学博士生招生
+2/48
-
同济大学环境学院肖倩课题组招收博士研究生（资格审核制）
+1/41
-
香港城市大学生物物理/软物质材料课题组2026-2027年招博士生1名
+2/22
-
评职称需要，卖发明专利了:D有技术储备需求的设备厂商可联系
+1/19
-
山东大学-山东字节信息智能技术实验室2026年招聘博士后
+2/16
-
上海大学微电子学院杨军教授团队招收微电子/集成电路/材料/机械/微纳制造博士一名
+1/9
-
哈工大马樱招收2027级计算机类/集成电路类博士生
+1/9
-
孩子爱看的纪录片，远比刷短视频更有价值
+1/8
-
上海大学微电子学院杨军教授团队招收微电子/集成电路/材料/机械/微纳制造博士一名
+1/8
-
【科研诚信】武汉轻工大学何江玲3篇论文：涉嫌抄袭、成组数据重复、制造虚假显著
+1/5
-
上海大学微电子学院杨军教授团队招收微电子/集成电路/材料/机械/微纳制造博士一名
+1/4
-
电机设计与电力电子变换器优化及AI大模型本地部署，算力工作站配置方案
+1/4
-
论文投出去之前到底要不要先申请专利？被审稿周期坑过一次后的复盘
+1/3
-
义翘神州推出全长p-Tau217蛋白，助力阿尔茨海默病下一代生物标志物检测开发
+1/3
-
吉林大学高压与超硬材料全国重点实验室李达教授课题组招收2027申请考核制博士
+1/2
-
结肠炎多维度分子解析：炎症‑免疫‑肠道屏障一体化检测策略
+1/1
-
科研助理招聘
+1/1
2楼2013-07-03 09:54:58
3楼2013-07-04 09:18:13
4楼2013-07-09 14:25:43
6楼2013-07-09 14:27:19
7楼2013-10-13 15:59:59
简单回复
jackyma5楼
2013-07-09 14:25
回复
