计算化学公社

 找回密码 Forget password
 注册 Register

使用gmxMMPBSA进行计算时轨迹文件的帧数是否会影响计算结果?

查看数: 1413 | 评论数: 9 | 收藏 Add to favorites 2
关灯 | 提示:支持键盘翻页<-左 右->
    组图打开中,请稍候......
发布时间: 2024-7-2 10:30

正文摘要:

各位老师好,目前我正在使用gmxMMPBSA进行计算,我的轨迹文件的帧数是5001帧,并且在我的mmpbsa.in文件中startframe和endframe分别设置为1和5001,但是在进行计算时总是计算到某一帧就报错(如下图),所以我想是不 ...

回复 Reply

12313 发表于 Post on 2024-7-3 11:36:00
低调的板凳 发表于 2024-7-3 10:46
可以直接在pb的脚本里调interval,也可以先用gmx trjconv把这些帧提出来,再计算pb。

好的,谢谢老师
低调的板凳 发表于 Post on 2024-7-3 10:46:16
12313 发表于 2024-7-2 15:42
老师,如果是要设置间隔20帧取一帧,是不是调整interval 这个参数啊?

可以直接在pb的脚本里调interval,也可以先用gmx trjconv把这些帧提出来,再计算pb。
12313 发表于 Post on 2024-7-2 16:21:27
student0618 发表于 2024-7-2 12:26
Sampling interval is how frequently a frame is selected from trajectory for calculation.
You can c ...

Thank you very much
12313 发表于 Post on 2024-7-2 15:42:10
低调的板凳 发表于 2024-7-2 14:38
5000帧的话在想要计算的平衡段平均取个200-300帧进行PB计算就行。假设500帧之后RMSD相对平衡,那就500-5000 ...

老师,如果是要设置间隔20帧取一帧,是不是调整interval 这个参数啊?
12313 发表于 Post on 2024-7-2 15:11:47
低调的板凳 发表于 2024-7-2 14:38
5000帧的话在想要计算的平衡段平均取个200-300帧进行PB计算就行。假设500帧之后RMSD相对平衡,那就500-5000 ...

感谢老师的回复
低调的板凳 发表于 Post on 2024-7-2 14:38:24
5000帧的话在想要计算的平衡段平均取个200-300帧进行PB计算就行。假设500帧之后RMSD相对平衡,那就500-5000帧间隔20帧取一帧计算就行。
student0618 发表于 Post on 2024-7-2 12:26:24
Sampling interval is how frequently a frame is selected from trajectory for calculation.
You can change the interval in mmpbsa.in (screenshot of input file taken from gmx_MMPBSA website )


Autocorrelation time depends on system. See https://valdes-tresanco-ms.github.io/gmx_MMPBSA/dev/api/?h=autoc on more details on why we want to use an uncorrelated dataset. original publication of mmpbsa.py (https://doi.org/10.1021/ct300418h) also discusses on it.
12313 发表于 Post on 2024-7-2 12:03:26
student0618 发表于 2024-7-2 11:00
Usually we need more than 100 frames to do MMPBSA, and the sampling interval should be longer than t ...

Thanks for your answer.By the way,What is the sampling interval and the autocorrelation time of the system?
Are these two parameters in certain files? e.g. mmpbsa.in
student0618 发表于 Post on 2024-7-2 11:00:08
本帖最后由 student0618 于 2024-7-2 11:08 编辑

Usually we need more than 100 frames to do MMPBSA, and the sampling interval should be longer than the autocorrelation time of the system.

Refer to the discussion in this thread on how to choose the frames for MMPBSA, which part of trajectory to use http://bbs.keinsci.com/forum.php ... 42&highlight=MMPBSA

For what happened exactly that caused the error you have, its better to check the other output files from the software, instead of just guessing.

手机版 Mobile version|北京科音自然科学研究中心 Beijing Kein Research Center for Natural Sciences|京公网安备 11010502035419号|计算化学公社 — 北京科音旗下高水平计算化学交流论坛 ( 京ICP备14038949号-1 )|网站地图

GMT+8, 2026-2-27 19:54 , Processed in 0.175598 second(s), 25 queries , Gzip On.

快速回复 返回顶部 返回列表 Return to list