terminal prolines

[cberti]

Hello. I am new to using the 2013 AMOEBA force field in OpenMM, but I have noticed a problem with my pdb files containing terminal prolines. I would try to run a one step simulation, and I would get the following error:

ValueError: No template found for residue 94 (PRO). The set of atoms is similar to GLH, but it is missing 1 atoms.

I realized that terminal prolines are supposed to be labeled as CPRO, so I modified the pdb file. Unfortunately, now when I run it, I receive a different error:

ValueError: No template found for residue 93 (ALA). The set of atoms matches ALA, but the bonds are different.

It appears that it's unhappy with the residue directly preceding the terminal proline now. Any advice? Thank you for your time.

CB

[peastman]

Sorry for the delay in replying to this! I've been on vacation for the last week.

PRO is the correct label for the residue in the PDB file, so don't change that. Otherwise, you'll just confuse the PDB reader and it won't know what bonds it's supposed to add.

It sounds like the problem is that your file is terminated in a slightly different way from what is defined in the AMOEBA force field. It expects a particular set of atoms, and a particular set of bonds between them. Can you post the section of the PDB file for that residue? Also, is it listed as the last residue in the chain, or is it followed by a separate terminator?

[cberti]

Hello,

I'm sorry for my own delayed response, hectic week. Nothing to apologize for though...

Would it be OK if I sent you some PDB files directly via e-mail? Thank you.

Christina Berti

----- Original Message -----
From: "peastman" notifications@github.com
To: "pandegroup/pdbfixer" pdbfixer@noreply.github.com
Cc: "cberti" cberti@purdue.edu
Sent: Tuesday, October 20, 2015 1:35:14 AM
Subject: Re: [pdbfixer] terminal prolines (#112)

Sorry for the delay in replying to this! I've been on vacation for the last week.

PRO is the correct label for the residue in the PDB file, so don't change that. Otherwise, you'll just confuse the PDB reader and it won't know what bonds it's supposed to add.

It sounds like the problem is that your file is terminated in a slightly different way from what is defined in the AMOEBA force field. It expects a particular set of atoms, and a particular set of bonds between them. Can you post the section of the PDB file for that residue? Also, is it listed as the last residue in the chain, or is it followed by a separate terminator?

—
Reply to this email directly or view it on GitHub .

[peastman]

Sure, that would be fine.

[cberti]

Thank you,

I am attaching a monomer of 2nsb.

Christina Berti

----- Original Message -----
From: "peastman" notifications@github.com
To: "pandegroup/pdbfixer" pdbfixer@noreply.github.com
Cc: "cberti" cberti@purdue.edu
Sent: Monday, October 26, 2015 1:30:53 PM
Subject: Re: [pdbfixer] terminal prolines (#112)

Sure, that would be fine.

—
Reply to this email directly or view it on GitHub .

[peastman]

I don't think github got the attachment. Can you email it to me directly? peastman at stanford edu.

[cberti]

That's fine, I'm attaching the same monomer.

Christina Berti

----- Original Message -----
From: "peastman" notifications@github.com
To: "pandegroup/pdbfixer" pdbfixer@noreply.github.com
Cc: "cberti" cberti@purdue.edu
Sent: Tuesday, October 27, 2015 3:52:16 PM
Subject: Re: [pdbfixer] terminal prolines (#112)

I don't think github got the attachment. Can you email it to me directly? peastman at stanford edu.

—
Reply to this email directly or view it on GitHub .

[peastman]

Got it, thanks. There's definitely a bug in the force field's template for CPRO. It claims that HA should be a carbon, which is clearly wrong. Let me investigate further.

[peastman]

The bug is present in the original Tinker force field definition (amoebapro13.prm). It specifies

biotype     648    HA      "C-Terminal PRO"                   55

But atom type 55 is defined as

atom         55   16    C     "Proline CB"                   6    12.011    4

So we should contact the Tinker developers about this.

[cberti]

Thank you! I had noticed this identical problem (failure to recognize terminal proline structures) when using the AMOEBA force field with OpenMM as well. Perhaps the AMOEBA developers would be the best to contact?

Christina Berti

----- Original Message -----
From: "peastman" notifications@github.com
To: "pandegroup/pdbfixer" pdbfixer@noreply.github.com
Cc: "cberti" cberti@purdue.edu
Sent: Tuesday, October 27, 2015 6:53:34 PM
Subject: Re: [pdbfixer] terminal prolines (#112)

The bug is present in the original Tinker force field definition (amoebapro13.prm). It specifies biotype 648 HA "C-Terminal PRO" 55

But atom type 55 is defined as atom 55 16 C "Proline CB" 6 12.011 4

So we should contact the Tinker developers about this.

—
Reply to this email directly or view it on GitHub .

[peastman]

I just sent an email about it. I'll let you know what I hear back.

[cberti]

Hello,

If it's more appropriate to address this question to the forum, I understand and apologize. But regarding Tinker's RESTRAIN-POSITION option in the .key file, would you happen to know the correct syntax for restraining all atoms except hydrogen atoms? Thank you.

Christina Berti

----- Original Message -----
From: "peastman" notifications@github.com
To: "pandegroup/pdbfixer" pdbfixer@noreply.github.com
Cc: "cberti" cberti@purdue.edu
Sent: Tuesday, October 27, 2015 7:23:09 PM
Subject: Re: [pdbfixer] terminal prolines (#112)

I just sent an email about it. I'll let you know what I hear back.

—
Reply to this email directly or view it on GitHub .

[peastman]

Sorry, I'm not too familiar with Tinker key file syntax.

I heard back from Jay Ponder, who confirmed this is a bug and he'll fix it in the next Tinker release. In the mean time, if you want to fix this yourself so you can move ahead, the error is in line 1182 of amoeba2013.xml:

   <Atom name="HA" type="55" />

Just change the type from 55 to 54.

[cberti]

Thank you so much! Within tinker-6.3.3, I fixed line 5717 of amoebapro13.prm found in the params directory. Thank you again!

Christina Berti

----- Original Message -----
From: "peastman" notifications@github.com
To: "pandegroup/pdbfixer" pdbfixer@noreply.github.com
Cc: "cberti" cberti@purdue.edu
Sent: Wednesday, October 28, 2015 2:29:07 PM
Subject: Re: [pdbfixer] terminal prolines (#112)

Sorry, I'm not too familiar with Tinker key file syntax.

I heard back from Jay Ponder, who confirmed this is a bug and he'll fix it in the next Tinker release. In the mean time, if you want to fix this yourself so you can move ahead, the error is in line 1182 of amoeba2013.xml:
< Atom name = " HA " type = " 55 " />

Just change the type from 55 to 54.

—
Reply to this email directly or view it on GitHub .

[aminemosbah]

hi
here is my error message ,
ValueError: No template found for residue 26 (ASN). The set of atoms is similar to GLN, but it is missing 2 atoms.
seems i have the same issue,i m not that accomplished at coding
would you please help me out,

[peastman]

@aminemosbah that isn't related. This issue is specifically about terminal prolines. If you're having a different problem could you open a new issue on it? Be sure to include all files needed to reproduce it.