Hydrogen bonds not getting detected

[nidhijbatra]

Hi I installed pycontact on 2 machines (Mac and Linux) by following the installation instructions in the tutorial and calculated the intra-protein contacts on trajectory. But I am getting H-bond percentage as 0 and no H-bonds are getting detected. However, I did the same calculations earlier (1.5 year ago) and it was working then. I am unable to figure out the problem. Do I have to install some other module for H-bond calculations? Please help regarding this.

[ZakkyZakster]

I'm experiencing the same problem when trying to see the H-bonds between ligand and the protein, was there a solution to this problem?

[RMCrean]

Hi both, Are you using .pdb files as your topologies perhaps?

This is likely causing issues for mdanalysis as explained here: https://docs.mdanalysis.org/2.0.0-dev0/documentation_pages/analysis/hydrogenbonds.html "It is highly recommended that a topology with bond information is used to generate the universe, e.g PSF, TPR, or PRMTOP files. This is the only method by which it can be guaranteed that donor-hydrogen pairs are correctly identified."

I had the same issue but then swapped from a .pdb as topology to a .prmtop (Amber topology) and now it works fine.

[chenxiang999]

@RMCrean hello, Do I need to manually modify the contents of the top file? How can I modify the contents to get an identifiable segid? Does the mdcrd file also need to be modified?

[RMCrean]

@chenxiang999 In my case I did not need to modify the topology or trajectory files in anyway. From memory (it was a little while ago), I used resid commands only. If you're using an Amber topology and trajectory a "segid" command might not work because Amber does not retain chain IDs (but you should be able to use a "resid" selection instead to get the same effect). Hope that helps.

[chenxiang999]

thanks！it works！ 

風 @.***

 

------------------ 原始邮件 ------------------ 发件人: "Rory @.>; 发送时间: 2023年10月26日(星期四) 晚上8:22 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [maxscheurer/pycontact] Hydrogen bonds not getting detected (#68)

@chenxiang999 In my case I did not need to modify the topology or trajectory files in anyway. From memory (it was a little while ago), I used resid commands only. If you're using an Amber topology and trajectory a "segid" command might not work because Amber does not retain chain IDs (but you should be able to use a "resid" selection instead to get the same effect). Hope that helps.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: @.***>